CN110022381A - A kind of load sharing method and device - Google Patents
A kind of load sharing method and device Download PDFInfo
- Publication number
- CN110022381A CN110022381A CN201910398893.2A CN201910398893A CN110022381A CN 110022381 A CN110022381 A CN 110022381A CN 201910398893 A CN201910398893 A CN 201910398893A CN 110022381 A CN110022381 A CN 110022381A
- Authority
- CN
- China
- Prior art keywords
- task
- mobile device
- local mobile
- edge cloud
- handled
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a kind of load sharing method and devices.The task sharing that this method is used to initiate local mobile device gives local mobile device and the processing of edge cloud, comprising: obtains the temporal sensitivity coefficient and computation complexity coefficient of task and the task that local mobile device is initiated；Calculate the task respective response time and energy consumption when local mobile device and edge cloud are handled；Cost function is obtained according to the temporal sensitivity coefficient of the task and computation complexity coefficient and the response time and energy consumption；When the cost function is greater than zero, the task is handled by edge cloud；When the cost function is less than or equal to zero, the task is handled by local mobile device.This method is shared for the different business feature selecting task by edge cloud and local mobile device, to improve the service quality of whole efficiency and task, and then improves user experience.
Description
Technical field
The present invention relates to fields of communication technology, and in particular to a kind of load sharing method and device.
Background technique
Traditional cloud computing is that all data are moved to cloud computing model, needs to occupy biggish Internet resources, Chang Zao
At congestion and when ductility it is poor.
For this purpose, related technical personnel propose edge cloud computing, i.e., operation is completed in the edge zone of data source.Edge cloud
Calculating belongs to a kind of distributing operation framework and moves application program, data information and the operation of service by Network Central Node
" fringe node " being sent on cellular logic is handled.
The rise of edge cloud computing is so that many business are put into edge cloud by user handles, but edge cloud is not always
There are enough bandwidth and computing resource come when meeting business transmission and processing.Moreover, because the feature of local service is different, clock synchronization
The requirement of ductility and arithmetic speed is different.The local service of different characteristics is placed on the processing of edge cloud, is not able to satisfy user
Demand.Such as the business of big data quantity, edge cloud computing is placed it in, can effectively play the advantage of edge cloud computing, at business
It manages high-efficient.However, the business more than but the frequency small for data volume, since clock synchronization Ductility Requirement is more sensitive, if put
In the processing of edge cloud, the time for transmitting data spends the time more than calculating itself, instead not as good as the efficiency for being placed on processing locality
It is high.
But the communications field is directed to the load that different business feature shares edge cloud and local not yet at present, leads to industry
The whole efficiency handled of being engaged in is not high, poor user experience.
Summary of the invention
For this purpose, the present invention provides a kind of load sharing method and device, to solve in the prior art due to lacking for not
It is shared by edge cloud with local mobile device with service feature and causes overall treatment efficiency low, the problem of poor user experience.
To achieve the goals above, the first aspect of the present invention provides a kind of load sharing method, is used for local movement
The task sharing that equipment is initiated gives local mobile device and the processing of edge cloud, comprising:
Obtain the temporal sensitivity coefficient and computation complexity coefficient of task and the task that local mobile device is initiated；
Calculate the task respective response time and energy consumption when local mobile device and edge cloud are handled；
It is obtained according to the temporal sensitivity coefficient of the task and computation complexity coefficient and the response time and energy consumption
Obtain cost function；
When the cost function is greater than zero, the task is handled by edge cloud；When the cost function is less than or equal to
When zero, the task is handled by local mobile device.
Wherein, the temporal sensitivity coefficient lambda of the taski tWith computation complexity coefficient lambdai eIt has the property that
And
Wherein, the response time that the task is handled in local mobile deviceAre as follows:
Wherein, i indicates i-th task, WiFor the periodicity for completing the CPU that i-th task needs, fi lIt is set for local movement
Revolution CPU per second when preparation goes out i-th task；
The response time that the task is handled in edge cloud are as follows:
Wherein,It is placed on the processing time that cloud processing in edge needs for i-th task,It is i-th task from local
Transmission time needed for mobile device is transferred to edge cloud,This is transferred to from edge cloud for the output result of i-th task
Transmission time needed for ground mobile device,The data volume of i-th task is sent to edge cloud for local mobile device,For
Edge cloud sends the output result data amount of i-th task to local mobile device,For the data volume for sending i-th task
Shared bandwidth,For bandwidth shared by i-th task computation result output data quantity, Pi inFor the sheet for initiating i-th task
The transmission power of ground mobile device, Pi outTo send the power that i-th task exports result from edge cloud to local mobile device,
HiFor channel gain, W0For Background Noise Power.
Wherein, the energy consumption that the task is handled in local mobile deviceAre as follows:
Wherein, Pi lThe power of i-th required by task, T are completed for local mobile devicei lI-th is completed for local mobile device
The response time of item task, α and β are constant, fi lRevolution CPU per second when issuing i-th task for local mobile device, WiFor
Complete the cpu cycle number of i-th task needs；
The energy consumption that the task is handled in edge cloudAre as follows:
Wherein, Ti tinFor i-th task from local mobile device be transferred to edge cloud needed for transmission time, Ti touT is
The output result of i tasks from edge cloud be transferred to local mobile device needed for transmission time, Pi ltAnd Pi lrIt refers respectively to send out
It send power and receives power,The data volume of i-th task is sent to edge cloud for local mobile device,For edge cloud to
Local mobile device sends the output result data amount of i-th task,To send band shared by the data volume of i-th task
Width,For bandwidth shared by i-th task computation result output data quantity, Pi inIt is set to initiate the local movement of i-th task
Standby transmission power, Pi outTo send the power that i-th task exports result, H from edge cloud to local mobile deviceiFor channel
Gain, W0For Background Noise Power.
Wherein, the cost function ViIt is obtained by following formula:
Wherein,For the temporal sensitivity coefficient of i-th task, Ti lIt is handled for i-th task in local mobile device
Response time, Ti rFor the response time that i-th task is handled in edge cloud, λi eFor the computation complexity coefficient of i-th task,For the energy consumption that i-th task is handled in local mobile device,For the energy consumption that i-th task is handled in edge cloud, Xi∈
{0,1}。
Wherein, when the cost function is greater than zero, the task is sent to institute through base station by the local mobile device
The processing of edge cloud is stated, processing result is sent to the local mobile device through the base station by the edge cloud.
The second aspect of the present invention provides a kind of load balancing device, the task sharing for initiating local mobile device
It is handled to local mobile device and edge cloud, comprising:
Acquiring unit, by obtain task and the task that local mobile device is initiated temporal sensitivity coefficient and based on
Calculate complexity factor；
First computing unit, when for calculating respective response when the task is handled by local mobile device and edge cloud
Between；
Second computing unit, for calculating respective energy consumption when the task is handled by local mobile device and edge cloud；
Third computing unit, for according to the temporal sensitivity coefficient and computation complexity coefficient of the task and described
Response time and energy consumption obtain cost function；
Judging unit, for transferring to edge cloud to handle the task when the cost function is greater than zero；When the generation
When valence function is less than or equal to zero, local mobile device is transferred to handle the task.
Wherein, first computing unit obtains the response time that the task is handled in local mobile deviceAre as follows:
Wherein, i indicates i-th task, WiFor the periodicity for completing the CPU that the task needs, fi lIt is set for local movement
Revolution CPU per second when preparation goes out i-th task；
The response time that the task is handled in edge cloud are as follows:
Wherein,It is placed on the processing time that cloud processing in edge needs for i-th task,It is i-th task from local
Transmission time needed for mobile device is transferred to edge cloud,This is transferred to from edge cloud for the output result of i-th task
Transmission time needed for ground mobile device,The data volume of i-th task is sent to edge cloud for local mobile device,For
Edge cloud sends the output result data amount of i-th task to local mobile device,For the data volume for sending i-th task
Shared bandwidth,For bandwidth shared by i-th task computation result output data quantity, Pi inFor the sheet for initiating i-th task
The transmission power of ground mobile device, Pi outTo send the power that i-th task exports result from edge cloud to local mobile device,
HiFor channel gain, W0For Background Noise Power.
Wherein, second computing unit obtains the energy consumption that the task is handled in local mobile deviceAre as follows:
Wherein, Pi lThe power of i-th required by task, T are completed for local mobile devicei lI-th is completed for local mobile device
The response time of item task, α and β are constant, fi lRevolution CPU per second when issuing i-th task for local mobile device, WiFor
Complete the cpu cycle number of i-th task needs；
The energy consumption that the task is handled in edge cloudAre as follows:
Wherein, Ti tinFor i-th task from local mobile device be transferred to edge cloud needed for transmission time, Ti toutIt is
The output result of i tasks from edge cloud be transferred to local mobile device needed for transmission time, Pi ltAnd Pi lrIt refers respectively to send out
It send power and receives power,The data volume of i-th task is sent to edge cloud for local mobile device,For edge cloud to
Local mobile device sends the output result data amount of i-th task,To send band shared by the data volume of i-th task
Width,For bandwidth shared by i-th task computation result output data quantity, Pi inIt is set to initiate the local movement of i-th task
Standby transmission power, Pi outTo send the power that i-th task exports result, H from edge cloud to local mobile deviceiFor channel
Gain, W0For Background Noise Power.
Wherein, the third computing unit, which is obtained, obtains the cost function V by following formulai,
Wherein,For the temporal sensitivity coefficient of i-th task, λi eFor the computation complexity coefficient of i-th task, and
And temporal sensitivity coefficientWith computation complexity coefficientHaveAndTi lAppoint for i-th
It is engaged in the response time handled in local mobile device, Ti rFor the response time that i-th task is handled in edge cloud,It is i-th
The energy consumption that task is handled in local mobile device,For the energy consumption that i-th task is handled in edge cloud, Xi∈{0,1}。
The present invention has the advantage that
Load sharing method provided by the invention, by the temporal sensitivity coefficient of task, computation complexity coefficient and
The task respective response time and energy consumption when being handled by local mobile device and edge cloud, obtain the cost letter of completion task
Number, when cost function is greater than zero, the task is handled by edge cloud；It is described when the cost function is less than or equal to zero
Task is handled by local mobile device, i.e., for the different business feature selecting task by edge cloud and local mobile device point
Load to improve the service quality of whole efficiency and task, and then improves user experience.
Detailed description of the invention
The drawings are intended to provide a further understanding of the invention, and constitutes part of specification, with following tool
Body embodiment is used to explain the present invention together, but is not construed as limiting the invention.
Fig. 1 is the schematic diagram that the embodiment of the present invention implements that load sharing method uses system；
Fig. 2 is a kind of flow chart of load sharing method provided in an embodiment of the present invention；
Fig. 3 is the functional block diagram of load balancing device provided in an embodiment of the present invention.
In the accompanying drawings:
11: local mobile device 12: base station
13: edge cloud 31: acquiring unit
32: the first computing unit, 33: the second computing unit
34: third computing unit 35: judging unit
Specific embodiment
Below in conjunction with attached drawing, detailed description of the preferred embodiments.It should be understood that this place is retouched
The specific embodiment stated is merely to illustrate and explain the present invention, and is not intended to restrict the invention.
The present embodiment provides a kind of load sharing methods.What the load sharing method was used to initiate local mobile device appoints
Business is shared to local mobile device and the processing of edge cloud.As shown in Figure 1, multiple local mobile devices 11 pass through wireless network links
It is connect with base station 12, base station 12 is connect by wirelessly or non-wirelessly link with edge cloud 13, and in other words, local mobile device 11 passes through
Base station 12 is communicated with edge cloud 13.Moreover, edge cloud 13 can be communicated to connect with multiple base stations 12, each base station 12 with it is multiple
Local mobile device 11 communicates to connect.
As shown in Fig. 2, load sharing method the following steps are included:
Step S1 obtains temporal sensitivity coefficient and the computation complexity system of task and task that local mobile device is initiated
Number.
Task is initiated by local mobile device, while the task of acquisition, obtains the characteristic coefficient of the task, i.e. the time is quick
Sensitivity coefficientWith computation complexity coefficient lambdai e。
In the present embodiment, temporal sensitivity coefficientWith computation complexity coefficient lambdai eIt has the property that
And
If temporal sensitivity coefficientIt is bigger, then it represents that requirement of the service quality of the task to time delay instantaneity is higher；
If computation complexity coefficient lambdai eIt is bigger, then it represents that the task data amount is bigger, and the energy for calculating consumption is higher.
Step S2, calculating task respective response time and energy consumption when local mobile device and edge cloud are handled.
In step s 2, the response time T that i-th task is handled in local mobile devicei lIt is obtained by formula (1):
Wherein, i indicates i-th task, WiTo complete the central processing unit (Central that i-th task needs
Processing Unit, abbreviation CPU) periodicity, fi lCPU is per second when issuing i-th task for local mobile device turns
Number.
If the response time T that i-th task is handled in edge cloudi rIt is obtained by formula (2):
Wherein,It is placed on the processing time that cloud processing in edge needs for i-th task,It is i-th task from local
Transmission time needed for mobile device is transferred to edge cloud,This is transferred to from edge cloud for the output result of i-th task
Transmission time needed for ground mobile device,The data volume of i-th task is sent to edge cloud for local mobile device,For
Edge cloud sends the output result data amount of i-th task to local mobile device,For the data volume for sending i-th task
Shared bandwidth,For bandwidth shared by i-th task computation result output data quantity, Pi inFor the sheet for initiating i-th task
The transmission power of ground mobile device, Pi outTo send the power that i-th task exports result from edge cloud to local mobile device,
HiFor channel gain, W0For Background Noise Power.
The energy consumption that i-th task is handled in local mobile deviceIt is obtained by formula (3):
Wherein, Pi lThe power of i-th required by task, T are completed for local mobile devicei lI-th is completed for local mobile device
The response time of item task, α and β are constant, fi lRevolution CPU per second when issuing i-th task for local mobile device, WiFor
Complete the cpu cycle number of i-th task needs.
The energy consumption that i-th task is handled in edge cloudIt is obtained by formula (4):
Wherein, Pi ltAnd Pi lrIt refers respectively to the transmission power of local mobile device and receives power, Ti tinFor i-th task
From local mobile device be transferred to edge cloud needed for transmission time, Ti toutIt is passed for the output result of i-th task from edge cloud
Transmission time needed for the defeated mobile device to local.
Formula (5) and formula (6) are substituted into formula (4), the energy consumption that i-th task is handled in edge cloudAre as follows:
Wherein, Ti tinFor i-th task from local mobile device be transferred to edge cloud needed for transmission time, Ti toutIt is
The output result of i tasks from edge cloud be transferred to local mobile device needed for transmission time, Pi ltAnd Pi lrIt refers respectively to send out
It send power and receives power,The data volume of i-th task is sent to edge cloud for local mobile device,For edge cloud to
Local mobile device sends the output result data amount of i-th task,To send band shared by the data volume of i-th task
Width,For bandwidth shared by i-th task computation result output data quantity, Pi inIt is set to initiate the local movement of i-th task
Standby transmission power, Pi outTo send the power that i-th task exports result, H from edge cloud to local mobile deviceiFor channel
Gain, W0For Background Noise Power.
Step S3 is obtained according to the temporal sensitivity coefficient of task and computation complexity coefficient and response time and energy consumption
Cost function.
In step s3, the response time T of i-th task is obtained according to formula (1) and formula (2)iAre as follows:
Wherein, Xi∈ { 0,1 }, works as XiWhen=0, indicate that i-th task is handled in local mobile device；Work as XiWhen=1, table
Show i-th task in the processing of edge cloud.
The energy consumption E of i-th task is obtained according to formula (3) and formula (4)iAre as follows:
Work as XiWhen=0, indicate that i-th task is handled in local mobile device；Work as XiWhen=1, indicate i-th task on side
The processing of edge cloud.
Formula (8) and formula (9) are substituted into and calculate cost function (10):
Obtain cost function Vi:
Wherein,For the temporal sensitivity coefficient of i-th task,It is handled for i-th task in local mobile device
Response time, Ti rFor the response time that i-th task is handled in edge cloud, λi eFor the computation complexity coefficient of i-th task,For the energy consumption that i-th task is handled in local mobile device,For the energy consumption that i-th task is handled in edge cloud, Xi∈
{0,1}。
Work as Xi=0, then Vi=2；So only considering Xi=1 the case where.Work as Xi=1, then ViDepending on task characteristic coefficient and
The comparative situation that time and energy consumption are handled in local and edge cloud.
Cost function Vi> 0, then i-th task is in the processing of edge cloud；As cost function Vi≤ 0, then i-th task is at this
The processing of ground mobile device.
Step S4, when cost function is greater than zero, task is handled by edge cloud；When cost function is less than or equal to zero,
Task is handled by local mobile device.
Such as:
The first situation: it is small for data volume but to the higher task of time delay sensitivity,
V is calculatedi=-3.4, then i-th task is handled in local mobile device.
Second situation: it is big for data volume and to the lower task of time delay sensitivity,
V is calculatedi=0.9, then i-th task is in the processing of edge cloud.
The third situation: it is medium for data volume and to the higher task of time delay sensitivity,
V is calculatedi=1.25, then i-th task is in the processing of edge cloud.
Load sharing method provided in this embodiment, by the temporal sensitivity coefficient of task, computation complexity coefficient with
And the task respective response time and energy consumption when being handled by local mobile device and edge cloud, obtain the cost letter of completion task
Number, when cost function is greater than zero, task is handled by edge cloud；When cost function is less than or equal to zero, task is by locally moving
Dynamic equipment processing is shared for the different business feature selecting task by edge cloud and local mobile device, to improve whole
The service quality of body efficiency and task, and then improve user experience.
The present embodiment also provides a kind of load balancing device, and the task sharing for initiating local mobile device is to local
Mobile device and the processing of edge cloud, are based on system shown in FIG. 1.
As shown in figure 3, load balancing device includes:
Acquiring unit 31, for obtain task and the task that local mobile device is initiated temporal sensitivity coefficient and
Computation complexity coefficient.
In the present embodiment, temporal sensitivity coefficientWith computation complexity coefficientIt has the property that
And
If temporal sensitivity coefficientIt is bigger, then it represents that requirement of the service quality of the task to time delay instantaneity is higher；
If computation complexity coefficient lambdai eIt is bigger, then it represents that the task data amount is bigger, and the energy for calculating consumption is higher.
First computing unit 32, for calculating respective response when the task is handled by local mobile device and edge cloud
Time.
The response time T that i-th task is handled in local mobile devicei lIt is obtained by formula (1):
Wherein, i indicates i-th task, WiFor the periodicity for completing the central processor CPU that i-th task needs, fi lFor
Local mobile device issues revolution CPU per second when i-th task.
If the response time T that i-th task is handled in edge cloudi rIt is obtained by formula (2):
Wherein,It is placed on the processing time that cloud processing in edge needs for i-th task,It is i-th task from local
Transmission time needed for mobile device is transferred to edge cloud,This is transferred to from edge cloud for the output result of i-th task
Transmission time needed for ground mobile device,The data volume of i-th task is sent to edge cloud for local mobile device,For
Edge cloud sends the output result data amount of i-th task to local mobile device,For the data volume for sending i-th task
Shared bandwidth,For bandwidth shared by i-th task computation result output data quantity, Pi inFor the sheet for initiating i-th task
The transmission power of ground mobile device, Pi outTo send the power that i-th task exports result from edge cloud to local mobile device,
HiFor channel gain, W0For Background Noise Power.
Second computing unit 33, for calculating respective energy when the task is handled by local mobile device and edge cloud
Consumption.
The energy consumption that i-th task is handled in local mobile deviceIt is obtained by formula (3):
Wherein, Pi lThe power of i-th required by task, T are completed for local mobile devicei lI-th is completed for local mobile device
The response time of item task, α and β are constant, fi lRevolution CPU per second when issuing i-th task for local mobile device, WiFor
Complete the cpu cycle number of i-th task needs.
The energy consumption that i-th task is handled in edge cloudIt is obtained by formula (4):
Wherein, Pi ltAnd Pi lrIt refers respectively to the transmission power of local mobile device and receives power, Ti tinFor i-th task
From local mobile device be transferred to edge cloud needed for transmission time, Ti toutIt is passed for the output result of i-th task from edge cloud
Transmission time needed for the defeated mobile device to local.
Formula (5) and formula (6) are substituted into formula (4), the energy consumption that i-th task is handled in edge cloudAre as follows:
Wherein, Ti tinFor i-th task from local mobile device be transferred to edge cloud needed for transmission time, Ti toutIt is
The output result of i tasks from edge cloud be transferred to local mobile device needed for transmission time, Pi ltAnd Pi lrIt refers respectively to send out
It send power and receives power, di inThe data volume of i-th task is sent to edge cloud for local mobile device,For edge cloud to
Local mobile device sends the output result data amount of i-th task,To send band shared by the data volume of i-th task
Width,For bandwidth shared by i-th task computation result output data quantity, Pi inIt is set to initiate the local movement of i-th task
Standby transmission power, Pi outTo send the power that i-th task exports result, H from edge cloud to local mobile deviceiFor channel
Gain, W0For Background Noise Power.
Third computing unit 34, for according to the temporal sensitivity coefficient and computation complexity coefficient of the task and institute
It states the response time and energy consumption obtains cost function.
The response time T of i-th task is obtained according to formula (1) and formula (2)iAre as follows:
Wherein, Xi∈ { 0,1 }, works as XiWhen=0, indicate that i-th task is handled in local mobile device；Work as XiWhen=1, table
Show i-th task in the processing of edge cloud.
The energy consumption E of i-th task is obtained according to formula (3) and formula (4)iAre as follows:
Work as XiWhen=0, indicate that i-th task is handled in local mobile device；Work as XiWhen=1, indicate i-th task on side
The processing of edge cloud.
Formula (8) and formula (9) are substituted into and calculate cost function (10):
Obtain cost function Vi:
Wherein, λi tFor the temporal sensitivity coefficient of i-th task, Ti lIt is handled for i-th task in local mobile device
Response time, Ti rFor the response time that i-th task is handled in edge cloud, λi eFor the computation complexity coefficient of i-th task,For the energy consumption that i-th task is handled in local mobile device,For the energy consumption that i-th task is handled in edge cloud, Xi∈
{0,1}。
Judging unit 35, for transferring to edge cloud to handle the task when the cost function is greater than zero；When described
When cost function is less than or equal to zero, local mobile device is transferred to handle the task.
Work as Xi=0, then Vi=2；So only considering Xi=1 the case where.Work as Xi=1, then ViDepending on task characteristic coefficient and
The comparative situation that time and energy consumption are handled in local and edge cloud.
Cost function Vi> 0, then i-th task is in the processing of edge cloud；As cost function Vi≤ 0, then i-th task is at this
The processing of ground mobile device.
Load balancing device provided in this embodiment obtains the temporal sensitivity coefficient of task by acquiring unit, calculates
Complexity factor, when calculating respective response when the task is handled by local mobile device and edge cloud by the first computing unit
Between, respective energy consumption when which is handled by local mobile device and edge cloud is calculated by the second computing unit, is calculated by third
Unit obtains generation according to the temporal sensitivity coefficient and computation complexity coefficient of the task and the response time and energy consumption
Valence function, and judge that the task is still handled by edge cloud by local mobile device by judging unit, when cost function is greater than zero
When, task is handled by edge cloud；When cost function is less than or equal to zero, task is handled by local mobile device, i.e., for not
It is shared with the business feature selecting task by edge cloud and local mobile device, to improve the Service Quality of whole efficiency and task
Amount, and then improve user experience.
It is understood that the principle that embodiment of above is intended to be merely illustrative of the present and the exemplary implementation that uses
Mode, however the present invention is not limited thereto.For those skilled in the art, essence of the invention is not being departed from
In the case where mind and essence, various changes and modifications can be made therein, these variations and modifications are also considered as protection scope of the present invention.
Claims (10)
1. a kind of load sharing method, the task sharing for initiating local mobile device gives local mobile device and edge cloud
Processing characterized by comprising
Obtain the temporal sensitivity coefficient and computation complexity coefficient of task and the task that local mobile device is initiated；
Calculate the task respective response time and energy consumption when local mobile device and edge cloud are handled；
Generation is obtained according to the temporal sensitivity coefficient of the task and computation complexity coefficient and the response time and energy consumption
Valence function；
When the cost function is greater than zero, the task is handled by edge cloud；When the cost function is less than or equal to zero,
The task is handled by local mobile device.
2. load sharing method according to claim 1, which is characterized in that the temporal sensitivity coefficient of the taskWith
Computation complexity coefficientIt has the property that
And
3. load sharing method according to claim 2, which is characterized in that the task is handled in local mobile device
Response timeAre as follows:
Wherein, i indicates i-th task, WiFor the periodicity for completing the CPU that i-th task needs, fi lFor local mobile device hair
Revolution CPU per second when i-th task out；
The response time that the task is handled in edge cloud are as follows:
Wherein,It is placed on the processing time that cloud processing in edge needs for i-th task,It is i-th task from local movement
Transmission time needed for equipment is transferred to edge cloud,Output result for i-th task is transferred to local shifting from edge cloud
Transmission time needed for dynamic equipment,The data volume of i-th task is sent to edge cloud for local mobile device,For edge
Cloud sends the output result data amount of i-th task to local mobile device,To send shared by the data volume of i-th task
Bandwidth,For bandwidth shared by i-th task computation result output data quantity, Pi inIt is moved to initiate the local of i-th task
The transmission power of dynamic equipment, Pi outTo send the power that i-th task exports result, H from edge cloud to local mobile deviceiFor
Channel gain, W0For Background Noise Power.
4. load sharing method according to claim 3, which is characterized in that the task is handled in local mobile device
Energy consumptionAre as follows:
Wherein, Pi lThe power of i-th required by task, T are completed for local mobile devicei lI-th is completed for local mobile device to appoint
The response time of business, α and β are constant, fi lRevolution CPU per second when issuing i-th task for local mobile device, WiTo complete
The cpu cycle number that i-th task needs；
The energy consumption that the task is handled in edge cloudAre as follows:
Wherein, Ti tinFor i-th task from local mobile device be transferred to edge cloud needed for transmission time, Ti toutIt is i-th
The output result of task from edge cloud be transferred to local mobile device needed for transmission time, Pi ltAnd Pi lrIt refers respectively to send function
Rate and reception power,The data volume of i-th task is sent to edge cloud for local mobile device,It is edge cloud to local
Mobile device sends the output result data amount of i-th task,To send bandwidth shared by the data volume of i-th task,For bandwidth shared by i-th task computation result output data quantity, Pi inFor the local mobile device for initiating i-th task
Send power, Pi outTo send the power that i-th task exports result, H from edge cloud to local mobile deviceiFor channel gain,
W0For Background Noise Power.
5. load sharing method according to claim 4, which is characterized in that the cost function ViIt is obtained by following formula
:
Wherein,For the temporal sensitivity coefficient of i-th task, Ti lThe response handled for i-th task in local mobile device
Time, Ti rFor the response time that i-th task is handled in edge cloud,For the computation complexity coefficient of i-th task,For
The energy consumption that i-th task is handled in local mobile device,For the energy consumption that i-th task is handled in edge cloud, Xi∈{0,1}。
6. load sharing method described in -5 any one according to claim 1, which is characterized in that be greater than in the cost function
When zero, the task is sent to the edge cloud through base station and handled by the local mobile device, and the edge cloud ties processing
Fruit is sent to the local mobile device through the base station.
7. a kind of load balancing device, the task sharing for initiating local mobile device gives local mobile device and edge cloud
Processing characterized by comprising
Acquiring unit, for obtaining the temporal sensitivity coefficient of task and the task that local mobile device is initiated and calculating multiple
Miscellaneous degree coefficient；
First computing unit, for calculating the respective response time when task is handled by local mobile device and edge cloud；
Second computing unit, for calculating respective energy consumption when the task is handled by local mobile device and edge cloud；
Third computing unit, for according to the temporal sensitivity coefficient and computation complexity coefficient of the task and the response
Time and energy consumption obtain cost function；
Judging unit, for transferring to edge cloud to handle the task when the cost function is greater than zero；When the cost letter
When number is less than or equal to zero, local mobile device is transferred to handle the task.
8. load balancing device according to claim 7, which is characterized in that first computing unit obtains the task
In the response time of local mobile device processingAre as follows:
Wherein, i indicates i-th task, WiFor the periodicity for completing the CPU that the task needs, fi lFor local mobile device hair
Revolution CPU per second when i-th task out；
The response time that the task is handled in edge cloud are as follows:
Wherein,It is placed on the processing time that cloud processing in edge needs for i-th task,It is i-th task from local movement
Transmission time needed for equipment is transferred to edge cloud,Output result for i-th task is transferred to local shifting from edge cloud
Transmission time needed for dynamic equipment,The data volume of i-th task is sent to edge cloud for local mobile device,For edge
Cloud sends the output result data amount of i-th task to local mobile device,To send shared by the data volume of i-th task
Bandwidth,For bandwidth shared by i-th task computation result output data quantity, Pi inIt is moved to initiate the local of i-th task
The transmission power of dynamic equipment, Pi outTo send the power that i-th task exports result, H from edge cloud to local mobile deviceiFor
Channel gain, W0For Background Noise Power.
9. load balancing device according to claim 8, which is characterized in that second computing unit obtains the task
In the energy consumption of local mobile device processingAre as follows:
Wherein, Pi lThe power of i-th required by task, T are completed for local mobile devicei lI-th is completed for local mobile device to appoint
The response time of business, α and β are constant, fi lRevolution CPU per second when issuing i-th task for local mobile device, WiTo complete
The cpu cycle number that i-th task needs；
The energy consumption that the task is handled in edge cloudAre as follows:
Wherein, Ti tinFor i-th task from local mobile device be transferred to edge cloud needed for transmission time, Ti toutIt is i-th
The output result of task from edge cloud be transferred to local mobile device needed for transmission time, Pi ltAnd Pi lrIt refers respectively to send function
Rate and reception power,The data volume of i-th task is sent to edge cloud for local mobile device,It is edge cloud to local
Mobile device sends the output result data amount of i-th task,To send bandwidth shared by the data volume of i-th task,For bandwidth shared by i-th task computation result output data quantity, Pi inFor the local mobile device for initiating i-th task
Send power, Pi outTo send the power that i-th task exports result, H from edge cloud to local mobile deviceiFor channel gain,
W0For Background Noise Power.
10. load balancing device according to claim 9, which is characterized in that the third computing unit obtain by with
Lower formula obtains the cost function Vi,
Wherein,For the temporal sensitivity coefficient of i-th task,For the computation complexity coefficient of i-th task, moreover, when
Between sensitivity coefficientWith computation complexity coefficientHaveAndTi lIt is i-th task at this
The response time of ground mobile device processing, Ti rFor the response time that i-th task is handled in edge cloud,Exist for i-th task
The energy consumption of local mobile device processing,For the energy consumption that i-th task is handled in edge cloud, Xi∈{0,1}。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910398893.2A CN110022381A (en) | 2019-05-14 | 2019-05-14 | A kind of load sharing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910398893.2A CN110022381A (en) | 2019-05-14 | 2019-05-14 | A kind of load sharing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110022381A true CN110022381A (en) | 2019-07-16 |
Family
ID=67193691
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910398893.2A Pending CN110022381A (en) | 2019-05-14 | 2019-05-14 | A kind of load sharing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110022381A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110399210A (en) * | 2019-07-30 | 2019-11-01 | 中国联合网络通信集团有限公司 | Method for scheduling task and device based on edge cloud |
| CN111682973A (en) * | 2020-08-17 | 2020-09-18 | 烽火通信科技股份有限公司 | Method and system for arranging edge cloud |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107911478A (en) * | 2017-12-06 | 2018-04-13 | 武汉理工大学 | Multi-user based on chemical reaction optimization algorithm calculates discharging method and device |
| CN107995660A (en) * | 2017-12-18 | 2018-05-04 | 重庆邮电大学 | Support Joint Task scheduling and the resource allocation methods of D2D- Edge Servers unloading |
| CN108920279A (en) * | 2018-07-13 | 2018-11-30 | 哈尔滨工业大学 | A kind of mobile edge calculations task discharging method under multi-user scene |
| CN109240818A (en) * | 2018-09-04 | 2019-01-18 | 中南大学 | Task discharging method based on user experience in a kind of edge calculations network |
| EP3457664A1 (en) * | 2017-09-14 | 2019-03-20 | Deutsche Telekom AG | Method and system for finding a next edge cloud for a mobile user |
-
2019
- 2019-05-14 CN CN201910398893.2A patent/CN110022381A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3457664A1 (en) * | 2017-09-14 | 2019-03-20 | Deutsche Telekom AG | Method and system for finding a next edge cloud for a mobile user |
| CN107911478A (en) * | 2017-12-06 | 2018-04-13 | 武汉理工大学 | Multi-user based on chemical reaction optimization algorithm calculates discharging method and device |
| CN107995660A (en) * | 2017-12-18 | 2018-05-04 | 重庆邮电大学 | Support Joint Task scheduling and the resource allocation methods of D2D- Edge Servers unloading |
| CN108920279A (en) * | 2018-07-13 | 2018-11-30 | 哈尔滨工业大学 | A kind of mobile edge calculations task discharging method under multi-user scene |
| CN109240818A (en) * | 2018-09-04 | 2019-01-18 | 中南大学 | Task discharging method based on user experience in a kind of edge calculations network |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110399210A (en) * | 2019-07-30 | 2019-11-01 | 中国联合网络通信集团有限公司 | Method for scheduling task and device based on edge cloud |
| CN110399210B (en) * | 2019-07-30 | 2021-10-01 | 中国联合网络通信集团有限公司 | Task scheduling method and device based on edge cloud |
| CN111682973A (en) * | 2020-08-17 | 2020-09-18 | 烽火通信科技股份有限公司 | Method and system for arranging edge cloud |
| CN111682973B (en) * | 2020-08-17 | 2020-11-13 | 烽火通信科技股份有限公司 | Method and system for arranging edge cloud |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Exploiting massive D2D collaboration for energy-efficient mobile edge computing | |
| Liang et al. | Joint resource allocation and coordinated computation offloading for fog radio access networks | |
| Yang et al. | Multivessel computation offloading in maritime mobile edge computing network | |
| CN109618373A (en) | Link choosing method and relevant apparatus | |
| Chen et al. | Delay guaranteed energy-efficient computation offloading for industrial IoT in fog computing | |
| CN107249217A (en) | The Joint Task unloading of ad hoc mobile cloud network and resource allocation methods | |
| Xiao et al. | System delay optimization for mobile edge computing | |
| Abedin et al. | Fog load balancing for massive machine type communications: A game and transport theoretic approach | |
| CN105516441A (en) | Information processing method and device | |
| CN104901989A (en) | Field service providing system and method | |
| CN104899103A (en) | Energy-saving scheduling method and energy-saving scheduling device for mobile cloud computing terminal | |
| CN109547174A (en) | A kind of method, the network equipment and the UE of time configuration | |
| Li et al. | Energy efficient relay selection and resource allocation in D2D-enabled mobile edge computing | |
| CN114138373A (en) | Edge calculation task unloading method based on reinforcement learning | |
| CN110213327A (en) | A kind of resource regulating method based on edge calculations, apparatus and system | |
| CN110022381A (en) | A kind of load sharing method and device | |
| CN108009024A (en) | Distributed game task discharging method in Ad-hoc cloud environments | |
| CN110149401A (en) | It is a kind of for optimizing the method and system of edge calculations task | |
| Zhang et al. | Buffer-aided device-to-device communication: Opportunities and challenges | |
| Nguyen et al. | Optimization of resource allocation for underlay device-to-device communications in cellular networks | |
| CN103581913B (en) | Cooperation transmission method and device in a kind of heterogeneous network | |
| Ravi et al. | Mobility managed energy efficient Android mobile devices using cloudlet | |
| Su et al. | Mobile cloud with smart offloading system | |
| CN110190982A (en) | It is a kind of based on fair time non-orthogonal multiple access the edge calculations time and energy consumption optimization method | |
| CN107835043A (en) | A kind of method of information transfer outage probability in rapid evaluation wireless power communication |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190716 |
|
| RJ01 | Rejection of invention patent application after publication |