CN110069325B - Task classification-based mobile edge computing task scheduling method - Google Patents
Task classification-based mobile edge computing task scheduling method Download PDFInfo
- Publication number
- CN110069325B CN110069325B CN201811029859.XA CN201811029859A CN110069325B CN 110069325 B CN110069325 B CN 110069325B CN 201811029859 A CN201811029859 A CN 201811029859A CN 110069325 B CN110069325 B CN 110069325B
- Authority
- CN
- China
- Prior art keywords
- task
- type
- processing
- task type
- tasks
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 48
- 238000005265 energy consumption Methods 0.000 claims abstract description 12
- 230000006870 function Effects 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 11
- 108090000623 proteins and genes Proteins 0.000 claims description 10
- 230000015654 memory Effects 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 6
- 238000004146 energy storage Methods 0.000 claims description 4
- 230000003993 interaction Effects 0.000 claims description 4
- 238000011160 research Methods 0.000 abstract description 2
- 238000004364 calculation method Methods 0.000 description 7
- 230000035772 mutation Effects 0.000 description 4
- 238000005457 optimization Methods 0.000 description 3
- 230000001413 cellular effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a task classification-based mobile edge computing task scheduling method, which aims to solve the problem that in the existing computing unloading research of mobile edge computing, different characteristics of different types of tasks are not considered, and the task cannot obtain the optimal delay characteristic meeting the task characteristics of the task; unloading each task by adopting a BOPTUD algorithm based on the optimal task parameter of each task type; the time delay of task processing can be obviously reduced, and the number of tasks processed by the mobile terminal can be increased under the condition that the energy consumption of the mobile terminal is the same.
Description
Technical Field
The invention belongs to the field of mobile edge computing, and particularly relates to a task scheduling technology based on mobile edge computing.
Background
As 4G technology matures, more and more mobile terminal users become accustomed to watching network video through diversified mobile terminals, and this new trend presents huge challenges to both wireless network operators and network video providers. While video service consumes a lot of wireless resources and bandwidth, network video providers want to provide differentiated service of their VIP users under intense competition, and at present, networks cannot provide differentiated service based on wireless base stations. As the 5G era is coming, the above requirements will be increasingly highlighted.
The above traffic demands are well met by the proposed mobile edge computing technique. Mobile Edge Computing (Mobile Edge Computing) is to deploy some servers at the Edge close to the Mobile end, so that some tasks that would have been handed to a data center far away before to process Computing can be processed in the server close to the Mobile end, thereby saving precious bandwidth resources in the network and reducing the communication load of the network, and therefore greatly reducing the processing delay of the tasks and improving the user experience.
In the mobile edge calculation, the calculation unloading is a very worth discussing problem. Since the energy of the mobile terminal is limited and the task processing delay expected by the mobile terminal has certain requirements for the task requirements, it is necessary to design an appropriate method for decision making to calculate and process tasks in the mobile terminal while considering the task processing delay and the energy consumed by the mobile terminal. Meanwhile, considering that different tasks have different characteristics, for differences among different tasks, the differences among the tasks need to be taken into account when designing a corresponding calculation unloading method.
Disclosure of Invention
In order to solve the problem that different characteristics of different types of tasks cannot be considered in the existing calculation unloading research of the mobile edge calculation, and the task cannot obtain the optimal delay characteristic meeting the task characteristics of the task, the invention provides a mobile edge calculation task scheduling method based on task classification.
The technical scheme adopted by the invention is as follows: the task classification-based mobile edge computing task scheduling method comprises the following steps: dividing the task into three task types according to task time delay requirements and task calculated amount; then determining initial task quantity parameters of each task type to be processed in a task waiting sequence; optimizing the task quantity parameters corresponding to each task type to obtain an optimal solution; and performing task offloading decision based on the optimal solution.
Further, an NTPO algorithm is adopted to optimize the task quantity parameters corresponding to each task type to obtain an optimal solution.
Further, optimizing the task quantity parameters corresponding to each task type by adopting an NTPO algorithm to obtain an optimal solution; the method specifically comprises the following steps:
s1, taking each task quantity parameter as an individual, and then coding the individual;
s2, generating a series of individuals meeting the task quantity parameter constraint condition;
s3, crossing the individuals to generate a plurality of filial generation individuals;
s4, if a certain child individual meets the constraint condition of the task quantity parameter, the new individual is reserved, and the total number of the child individuals is added with 1; otherwise, removing the individual;
s5, if the total number of the current filial generation individuals is more than 2 times of the initial population number or the total number of the current filial generation individuals reaches the upper limit of gene cross exchange; step S6 is executed; otherwise, returning to the step S3;
s6, selecting the optimal offspring individuals with the same number as the initial population by adopting a fitness function;
s7, if the gene segments of the individuals are mutated, executing the step S8 if the mutated filial generation individuals do not meet the constraint conditions of the task quantity parameters; otherwise, returning to the step S3;
and S8, executing a plurality of gene crossing operations, and screening out the best individual by adopting a fitness function.
Further, the task types are divided into three categories, including: a first task type, a second task type, and a third task type.
Further, the task amount parameter constraint conditions in step S2 are: further, the task amount parameter constraint conditions in step S2 are:
a>b>c
wherein, T3、T2、T1An upper delay requirement limit, T, tolerable for the third task type3、T2、T1An upper delay requirement limit, T, tolerable for the second task type3、T2、T1The upper delay requirement limit which can be tolerated by the third task type is the upper delay requirement limit which can be tolerated by the first task type, ti1、ti2、ti3General expression for latency that the first task type will bring to other tasks, ti1、ti2、ti3General expression for latency that the second task type will bring to other tasks, ti1、ti2、ti3A represents the task quantity parameter corresponding to the first type of task, b represents the task quantity parameter corresponding to the second type of task, c represents the task quantity parameter corresponding to the third type of task, ai，bi，ciIs a coefficient with a value of 0 or 1.
Further, the fitness function in step S6 is:
and further, performing task unloading decision on each task by adopting a BOPTUD algorithm based on the optimal solution.
Further, based on the optimal solution, adopting a BOPTUD algorithm to carry out task unloading decision on each task; the method specifically comprises the following steps:
t1, inputting the maximum length L of a task waiting sequence and calculating the optimal task quantity parameter of each task type by an NTPO algorithm;
t2, traversing the tasks in the task waiting sequence;
t3, determining the processing priority of each task type, and processing each task according to the task quantity parameters of each task type from high to low according to the priority;
t4, for each task, triggering one interaction with the mobile edge server to obtain the CPU utilization rate and the memory utilization rate of the server; calculating a load condition balance value of the server according to the acquired CPU utilization rate and memory utilization rate of the server;
t5, for a certain task, comparing the time delay of the server with the lightest load for processing the task with the time delay required by the local mobile terminal, and selecting the server with smaller time delay to process the task;
t6, calculating energy consumption required by the mobile terminal in each task processing process;
and T7, calculating the accumulated energy consumption of the processed tasks of the mobile terminal, if the accumulated energy consumption exceeds the maximum energy storage of the mobile terminal, terminating the algorithm, and otherwise, continuously processing the next task.
Further, step T3 is specifically: specifically, the priority is, in order from high to low: a first task type, a second task type, a third task type; processing each task according to the task quantity parameter of each task type from high priority to low priority, which specifically comprises the following steps:
a1, judging whether the first task type has task quantity parameters which are more than or equal to the first task type, if so, processing the task quantity parameters which are corresponding to the first task type, and then processing the second task type; otherwise, processing all tasks in the first task type, and then processing the second task type;
a2, judging whether the second task type has task quantity parameters which are more than or equal to the task quantity parameters corresponding to the second task type, if so, processing the task quantity parameters corresponding to the second task type, and then processing a third task type; otherwise, processing all tasks in the second task type, and then processing a third task type;
a3, judging whether the third task type has task quantity parameters which are more than or equal to the third task type, if so, processing the task quantity parameters which correspond to the third task type; otherwise, all tasks in the third task type are processed.
The invention has the beneficial effects that: the invention relates to a task classification-based mobile edge computing task scheduling method, which comprises the steps of firstly determining task types, giving initial task quantity parameters of each task type, and computing optimal task quantity parameters corresponding to each task type through an NTPO algorithm; unloading each task by adopting a BOPTUD algorithm based on the optimal task parameter of each task type; the method can obviously reduce the time delay of task processing, and improve the number of tasks processed by the mobile terminal under the condition that the energy consumption of the mobile terminal is the same.
Drawings
FIG. 1 is a schematic diagram of a computational offload model in moving edge computation implemented by the present invention.
Fig. 2 is a flowchart of the task number parameter setting optimization algorithm provided by the present invention.
Fig. 3 is a flowchart of a task scheduling algorithm provided by the present invention.
Detailed Description
In order to facilitate the understanding of the technical contents of the present invention by those skilled in the art, the present invention will be further explained with reference to the accompanying drawings.
Fig. 1 is a schematic diagram of a mobile terminal and a mobile edge server in a cellular wireless communication system, which includes a base station, a mobile terminal (e.g., a mobile phone, a tablet computer, etc.), and a mobile edge server.
A base station: the mobile terminal wireless network establishes connection with the base station, and the base station establishes connection with the server through the wired network, but data is supposed to arrive at the server after being transmitted from the mobile terminal to the base station (ignoring delay from the base station to the server).
A mobile terminal: the mobile terminal can only directly interact with the base station through the wireless network, and although several mobile terminals are illustrated in fig. 1, this is only for illustrating the diversity of the mobile terminals, and the present invention is only for one mobile terminal. The mobile terminal is the main body for executing the task unloading decision method.
Moving the edge server: it is assumed that the server does not have a data forwarding function, that is, the server only has a task to come and then performs task processing.
Network: in the figure, the straight line between the server and the base station represents a wired network, and the lightning symbol between the base station and the mobile terminal represents a cellular wireless network.
The technical scheme of the invention is as follows: a Task scheduling method Based on Task classification for mobile edge computing comprises a Number of Task Parameter Optimization algorithms (NTPO) corresponding to each Task type in a Task waiting sequence and a Task Unloading Decision algorithm (BOPTUD) Based on Optimization parameters.
In the present invention, the generation scenario of the task may be exemplified by a future intelligent manufacturing plant. Tasks can be divided into three categories, wherein the first category is tasks with highest delay requirement and small calculated amount, such as monitoring tasks of temperature, humidity, dust concentration in a workshop and the like; the second kind is a task with high delay requirement and high computation amount, such as a voice image task; a third category is tasks with high latency requirements but relatively small computational effort, such as indoor positioning-type tasks.
The invention is used in the mobile terminal, namely, each mobile terminal has different optimal task quantity parameters a, b and c due to different processing tasks, and the NTPO algorithm can generate different optimal task quantity parameters a, b and c for different mobile terminals. The NTPO algorithm firstly generates an initial population (i.e. a series of initial solutions meeting the constraint condition) by using the basic idea of the genetic algorithm, then the population individuals are continuously crossed and varied to generate offspring, the constraint condition and the fitness function are used for screening and leaving the optimal individuals with the same number as the initial population (i.e. ensuring that the offspring is continuously evolved and the population number is kept unchanged) until the individuals cannot be varied (i.e. the parameters are slightly changed so as not to meet the constraint condition), and the fitness function is used for screening the optimal individuals (i.e. the optimal solutions). As shown in fig. 2, the method specifically includes the following steps:
s1, encoding the three task quantity parameters a, b and c relating to the waiting task sequence into a single unit, such as (5, 3, 2) indicating that a is 5, b is 3 and c is 2
S2, generating an initial population (composed of a plurality of individuals similar to the individuals in S1) meeting the constraint condition.
S3, cross-exchanging genes among individuals (for example, a parameter in the individual A and B and c parameters in the individual B are recombined to obtain a new individual) to obtain a new filial generation individual;
s4, if a certain child individual meets the constraint condition of the task quantity parameter, the new individual is reserved, and the total number of the child individuals is added with 1;
s5, if the total number of the current filial generation individuals is more than 2 times of the initial population number or the total number of the current filial generation individuals reaches the upper limit of gene cross exchange; step S6 is executed; otherwise, returning to the step S3;
and S6, screening the best generation of individuals with the same number as the initial population by adopting a fitness function.
S7, carrying out mutation on the individual gene (namely, increasing the task quantity parameter), if the mutation does not accord with the constraint condition, then the mutation is abandoned, and the corresponding parameter value is restored to the parameter value before the mutation; then, step S8 is executed; otherwise, returning to the step S3;
if all individuals can not be mutated any more (namely, any one of the increased task quantity parameters can cause that the individuals do not accord with the constraint conditions), the individuals are exchanged for a plurality of generations in a cross way, each generation is screened by using the constraint conditions and the fitness function, the population number is kept unchanged, and finally the optimal individual is selected.
If the individual can be mutated again, then the step S3 is carried out, the NTPO algorithm screens the individual by using the constraint condition and the fitness function, and the fitness function is as follows:
the closer f (a, b, c) is to 0, the more excellent the corresponding set of task quantity parameters. The constraint conditions are as follows:
a>b>c
and S8, executing a plurality of gene crossing operations, and screening out the best individual by adopting a fitness function.
The BOPTUD algorithm processes tasks according to a task processing principle by judging the types of the tasks firstly, triggers interaction with surrounding servers once so as to know the utilization rate of CPUs and memories of the surrounding servers, calculates the load condition balance value of each server, picks out the server with the lightest load, and makes a decision to process the task on the corresponding server. As shown in fig. 3, the BOPTUD algorithm specifically includes the following steps:
t1, inputting the maximum length L of the waiting sequence, the numerical values of the parameters a, b and c and the maximum energy storage E of the mobile terminal;
t2, traversing the tasks in the task waiting sequence;
and T3, processing a tasks of the first task type at most to process a tasks of the second task type, and processing the tasks of the second task type if L tasks are not the first task type after the last processing of the tasks of the first task type. Then at most b tasks of the second task type are processed and then the third type tasks are processed, and then at most c tasks of the third type are processed and then the tasks of the first task type are processed.
And T4, when the number a or b or c is not reached, processing the next type of task according to the sequence, namely according to the sequence of the first task type, the second task type and the third task type.
T5, triggering one interaction with the mobile edge server for one task i each time to know the utilization rate of the CPU and the memory of the server to calculate the load condition balance value of the server;
F(Si)＝k1*C(Si)+k2*M(Si)。
wherein,C(Si) Indicating the utilization of the server CPU, M (S)i) Expressing the memory utilization, k, of the node servers1、k2Is a proportionality coefficient, and has a value range of [0, 1]In between, the sum of the two is 1, and here, the default value is 1/2.
T6, for task i, if it is processed in the server, selecting the one with the lightest load, predicting the time delay of processing task in the corresponding server and the time delay needed for processing at the local mobile terminal, and selecting the smallest one to make the corresponding selection (if the local time delay is smaller, then not uploading the processing, otherwise, unloading the processing to the server).
T7, calculating the energy consumption required by the mobile terminal in the process of processing the task i (whether the task is processed locally or at the server, the energy consumption of the mobile terminal in the process is calculated).
And T8, calculating the accumulated energy consumption sumE of all the tasks, if the accumulated energy consumption sumE of all the tasks exceeds the maximum energy storage E of the mobile terminal, terminating the algorithm, and otherwise, starting processing and considering the next task of the waiting sequence according to the convention traversed by the tasks.
It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (6)
1. The task classification-based mobile edge computing task scheduling method is characterized by comprising the following steps: dividing the task into three task types according to task time delay requirements and task calculated amount; then determining initial task quantity parameters of each task type to be processed in a task waiting sequence; optimizing the task quantity parameters corresponding to each task type to obtain an optimal solution; performing task unloading decision based on the optimal solution;
optimizing the task quantity parameters corresponding to each task type by adopting an NTPO algorithm to obtain an optimal solution; the method specifically comprises the following steps:
s1, taking each task quantity parameter as an individual, and then coding the individual;
s2, generating a series of individuals meeting the task quantity parameter constraint condition;
s3, crossing the individuals to generate a plurality of filial generation individuals;
s4, if a certain child individual meets the constraint condition of the task quantity parameter, the new individual is reserved, and the total number of the child individuals is added with 1; otherwise, removing the individual;
s5, if the total number of the current filial generation individuals is more than 2 times of the initial population number or the total number of the current filial generation individuals reaches the upper limit of gene cross exchange; step S6 is executed; otherwise, returning to the step S3;
s6, selecting the optimal offspring individuals with the same number as the initial population by adopting a fitness function; the fitness function in step S6 is:
wherein, T3、T2、T1An upper delay requirement limit, T, tolerable for the third task type3、T2、T1An upper delay requirement limit, T, tolerable for the second task type3、T2、T1The upper delay requirement limit which can be tolerated by the third task type is the upper delay requirement limit which can be tolerated by the first task type, ti1、ti2、ti3General expression for latency that the first task type will bring to other tasks, ti1、ti2、ti3General expression for latency that the second task type will bring to other tasks, ti1、ti2、ti3A represents the task quantity parameter corresponding to the first type of task, and b represents the task quantity parameter corresponding to the second type of taskC represents a task quantity parameter corresponding to the third type of task type, ai，bi，ciIs a coefficient with a value of 0 or 1;
s7, if the gene segments of the individuals are mutated, executing the step S8 if the mutated filial generation individuals do not meet the constraint conditions of the task quantity parameters; otherwise, returning to the step S3;
and S8, executing a plurality of gene crossing operations, and screening out the best individual by adopting a fitness function.
2. The task classification-based mobile edge computing task scheduling method of claim 1, wherein the task types are classified into three categories, comprising: a first task type, a second task type, and a third task type.
3. The task classification-based mobile edge computing task scheduling method of claim 2, wherein the task quantity parameter constraint condition in step S2 is:
4. the task classification-based mobile edge computing task scheduling method of claim 3, wherein a BOPTUD algorithm is used to make task offload decisions for each task based on the optimal solution.
5. The task classification-based mobile edge computing task scheduling method of claim 4, wherein a BOPTUD algorithm is used to make task offload decisions for each task based on an optimal solution; the method specifically comprises the following steps:
t1, inputting the maximum length L of a task waiting sequence and calculating the optimal task quantity parameter of each task type by an NTPO algorithm;
t2, traversing the tasks in the task waiting sequence;
t3, determining the processing priority of each task type, and processing each task according to the task quantity parameters of each task type from high to low according to the priority;
t4, for each task, triggering one interaction with the mobile edge server to obtain the CPU utilization rate and the memory utilization rate of the server; calculating a load condition balance value of the server according to the acquired CPU utilization rate and memory utilization rate of the server;
t5, for a certain task, comparing the time delay of the server with the lightest load for processing the task with the time delay required by the local mobile terminal, and selecting the server with smaller time delay to process the task;
t6, calculating energy consumption required by the mobile terminal in each task processing process;
and T7, calculating the accumulated energy consumption of the processed tasks of the mobile terminal, if the accumulated energy consumption exceeds the maximum energy storage of the mobile terminal, terminating the algorithm, and otherwise, continuously processing the next task.
6. The task classification-based mobile edge computing task scheduling method according to claim 5, wherein step T3 specifically comprises: specifically, the priority is, in order from high to low: a first task type, a second task type, a third task type; processing each task according to the task quantity parameter of each task type from high priority to low priority, which specifically comprises the following steps:
a1, judging whether the first task type has task quantity parameters which are more than or equal to the first task type, if so, processing the task quantity parameters which are corresponding to the first task type, and then processing the second task type; otherwise, processing all tasks in the first task type, and then processing the second task type;
a2, judging whether the second task type has task quantity parameters which are more than or equal to the task quantity parameters corresponding to the second task type, if so, processing the task quantity parameters corresponding to the second task type, and then processing a third task type; otherwise, processing all tasks in the second task type, and then processing a third task type;
a3, judging whether the third task type has task quantity parameters which are more than or equal to the third task type, if so, processing the task quantity parameters which correspond to the third task type; otherwise, all tasks in the third task type are processed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811029859.XA CN110069325B (en) | 2018-09-05 | 2018-09-05 | Task classification-based mobile edge computing task scheduling method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811029859.XA CN110069325B (en) | 2018-09-05 | 2018-09-05 | Task classification-based mobile edge computing task scheduling method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110069325A CN110069325A (en) | 2019-07-30 |
| CN110069325B true CN110069325B (en) | 2020-12-15 |
Family
ID=67365813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811029859.XA Expired - Fee Related CN110069325B (en) | 2018-09-05 | 2018-09-05 | Task classification-based mobile edge computing task scheduling method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110069325B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11864897B2 (en) | 2021-04-12 | 2024-01-09 | Toyota Research Institute, Inc. | Systems and methods for classifying user tasks as being system 1 tasks or system 2 tasks |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110489233A (en) * | 2019-08-15 | 2019-11-22 | 北京信息科技大学 | Equipment task unloading and cpu frequency modulation method and system based on mobile edge calculations |
| CN111148134B (en) * | 2019-12-19 | 2021-06-01 | 南京大学 | Multi-user multi-task unloading method based on mobile edge calculation |
| CN111459628B (en) * | 2020-03-12 | 2023-11-28 | 大庆市凯德信信息技术有限公司 | Spark platform task scheduling method based on improved quantum ant colony algorithm |
| CN111585816B (en) * | 2020-05-11 | 2022-07-01 | 重庆邮电大学 | Task unloading decision method based on adaptive genetic algorithm |
| CN112073452B (en) * | 2020-05-27 | 2021-04-27 | 河南工程学院 | Mobile edge computing task allocation method with effective energy and limited resources |
| CN111836284B (en) * | 2020-07-08 | 2022-04-05 | 重庆邮电大学 | Energy consumption optimization calculation and unloading method and system based on mobile edge calculation |
| CN111901435B (en) * | 2020-07-31 | 2021-09-17 | 南京航空航天大学 | Load-aware cloud-edge collaborative service deployment method |
| CN112083967B (en) * | 2020-08-18 | 2023-10-20 | 深圳供电局有限公司 | Cloud edge computing task unloading method, computer equipment and storage medium |
| CN111913855A (en) * | 2020-09-21 | 2020-11-10 | 北京百度网讯科技有限公司 | Method and device for determining target task calculation amount |
| CN112882809A (en) * | 2021-02-23 | 2021-06-01 | 国汽(北京)智能网联汽车研究院有限公司 | Method and device for determining computing terminal of driving task and computer equipment |
| CN112988275B (en) * | 2021-03-26 | 2022-10-14 | 河海大学 | Task perception-based mobile edge computing multi-user computing unloading method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106534333A (en) * | 2016-11-30 | 2017-03-22 | 北京邮电大学 | Bidirectional selection computing unloading method based on MEC and MCC |
| CN107249218A (en) * | 2017-06-05 | 2017-10-13 | 东南大学 | Radio Resource and the combined distributing method of cloud resource in a kind of MEC |
| CN107682443A (en) * | 2017-10-19 | 2018-02-09 | 北京工业大学 | Joint considers the efficient discharging method of the mobile edge calculations system-computed task of delay and energy expenditure |
| CN108319502A (en) * | 2018-02-06 | 2018-07-24 | 广东工业大学 | A kind of method and device of the D2D tasks distribution based on mobile edge calculations |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050192680A1 (en) * | 2004-02-27 | 2005-09-01 | Mark Cascia | System and method for optimizing global set points in a building environmental management system |
| US10944668B2 (en) * | 2017-02-27 | 2021-03-09 | Mavenir Networks, Inc. | System and method for supporting low latency applications in a cloud radio access network |
-
2018
- 2018-09-05 CN CN201811029859.XA patent/CN110069325B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106534333A (en) * | 2016-11-30 | 2017-03-22 | 北京邮电大学 | Bidirectional selection computing unloading method based on MEC and MCC |
| CN107249218A (en) * | 2017-06-05 | 2017-10-13 | 东南大学 | Radio Resource and the combined distributing method of cloud resource in a kind of MEC |
| CN107682443A (en) * | 2017-10-19 | 2018-02-09 | 北京工业大学 | Joint considers the efficient discharging method of the mobile edge calculations system-computed task of delay and energy expenditure |
| CN108319502A (en) * | 2018-02-06 | 2018-07-24 | 广东工业大学 | A kind of method and device of the D2D tasks distribution based on mobile edge calculations |
Non-Patent Citations (4)
| Title |
|---|
| A Task Assignment Algorithm Based on Particle Swarm Optimization and Simulated Annealing in Ad-hoc Mobile Cloud;Bonan Huang,Weiwei Xia,etc;《2017年第9届无线通信和信号处理国际会议》;20171211;1-6 * |
| 移动云计算下带能量约束的任务调度研究;黄航宇;《中国优秀硕士学位论文全文数据库 信息科技辑》;20180815;I137-2 * |
| 移动边缘计算（MEC）中任务协同调度策略;焦捷;《中国优秀硕士学位论文全文数据库 信息科技辑》;20180815;I136-248 * |
| 移动边缘计算任务卸载和基站关联协同决策问题研究;于博文，蒲凌君，谢玉婷，徐敬东，张建忠;《计算机研究与发展》;20180315;537-550 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11864897B2 (en) | 2021-04-12 | 2024-01-09 | Toyota Research Institute, Inc. | Systems and methods for classifying user tasks as being system 1 tasks or system 2 tasks |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110069325A (en) | 2019-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110069325B (en) | Task classification-based mobile edge computing task scheduling method | |
| Chen et al. | Energy-efficient task offloading and resource allocation via deep reinforcement learning for augmented reality in mobile edge networks | |
| CN111585816B (en) | Task unloading decision method based on adaptive genetic algorithm | |
| Sun et al. | Autonomous resource slicing for virtualized vehicular networks with D2D communications based on deep reinforcement learning | |
| Cui et al. | A novel offloading scheduling method for mobile application in mobile edge computing | |
| CN110493360A (en) | The mobile edge calculations discharging method of system energy consumption is reduced under multiserver | |
| CN108112037B (en) | Load balancing method based on fog calculation and cooperative communication network | |
| CN109756578B (en) | Low-delay task scheduling method for dynamic fog computing network | |
| WO2023040022A1 (en) | Computing and network collaboration-based distributed computation offloading method in random network | |
| CN105376157B (en) | A kind of method and apparatus of routed path selection | |
| CN104684095B (en) | Resource allocation methods based on genetic operation in a kind of heterogeneous network converged scene | |
| CN103888371B (en) | A kind of method and apparatus of virtual network mapping | |
| CN113220364A (en) | Task unloading method based on vehicle networking mobile edge computing system model | |
| CN110851277A (en) | Task scheduling strategy based on edge cloud cooperation in augmented reality scene | |
| CN112083967B (en) | Cloud edge computing task unloading method, computer equipment and storage medium | |
| CN112272102B (en) | Method and device for unloading and scheduling edge network service | |
| CN114785692A (en) | Virtual power plant aggregation regulation and control communication network flow balancing method and device | |
| CN113676357A (en) | Decision method for edge data processing in power internet of things and application thereof | |
| CN111580943B (en) | Task scheduling method for multi-hop unloading in low-delay edge calculation | |
| Wang et al. | Multi-objective joint optimization of communication-computation-caching resources in mobile edge computing | |
| Mu et al. | Stochastic learning for opportunistic peer-to-peer computation offloading in iot edge computing | |
| CN112084034A (en) | MCT scheduling method based on edge platform layer adjustment coefficient | |
| CN114615705B (en) | Single-user resource allocation strategy method based on 5G network | |
| Gao et al. | An integration of online learning and online control for green offloading in fog-assisted IoT systems | |
| CN114281544A (en) | Electric power task execution method and device based on edge calculation |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201215 Termination date: 20210905 |