CN107734558A - A kind of control of mobile edge calculations and resource regulating method based on multiserver - Google Patents
A kind of control of mobile edge calculations and resource regulating method based on multiserver Download PDFInfo
- Publication number
- CN107734558A CN107734558A CN201711011580.4A CN201711011580A CN107734558A CN 107734558 A CN107734558 A CN 107734558A CN 201711011580 A CN201711011580 A CN 201711011580A CN 107734558 A CN107734558 A CN 107734558A
- Authority
- CN
- China
- Prior art keywords
- base station
- task
- terminal
- data
- calculation
- 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
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000001105 regulatory effect Effects 0.000 title 1
- 239000011159 matrix material Substances 0.000 claims abstract description 32
- 230000005012 migration Effects 0.000 claims abstract description 27
- 238000013508 migration Methods 0.000 claims abstract description 27
- 230000005540 biological transmission Effects 0.000 claims abstract description 14
- 230000008569 process Effects 0.000 claims description 8
- 230000004044 response Effects 0.000 claims description 6
- 238000013459 approach Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 2
- 230000001934 delay Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 5
- 238000005265 energy consumption Methods 0.000 abstract description 4
- 238000013468 resource allocation Methods 0.000 abstract 2
- 238000010586 diagram Methods 0.000 description 6
- 230000009471 action Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 208000001613 Gambling Diseases 0.000 description 2
- 238000010295 mobile communication Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003190 augmentative effect Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000006386 memory function Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/08—Load balancing or load distribution
-
- 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]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
- G06F9/5088—Techniques for rebalancing the load in a distributed system involving task migration
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/50—Allocation or scheduling criteria for wireless resources
- H04W72/52—Allocation or scheduling criteria for wireless resources based on load
-
- 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)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (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 kind of resource allocation based on mobile edge calculations and base station service arrangement method, this method to include：When detect have calculating task in mobile terminal when, to intelligent base station send computation migration request；When lacking the calculating data needed for the task requests in the buffer unit of base station, to task data demand needed for network side transmission；Receive the required task data of network side return；According to the required task data received, calculation delay income and energy consumption income；Computation migration judgement matrix is obtained according to experience utility function；Matrix is adjudicated according to computation migration and carries out computation migration.The base station service arrangement scheme includes buffer unit, computing unit, obtains processing unit, transmitting element, can provide computing capability and data caching capabilities.Therefore, the method for the resource allocation based on MEC and base station service arrangement scheme, it can realize that terminal multitask, base station be multi-functional, the computation migration of target diversification.
Description
Technical Field
The invention relates to the technical field of mobile computing, in particular to a control and resource scheduling method based on mobile edge computing comprising multiple servers, and particularly relates to a control and resource scheduling method based on mobile edge computing between multiple mobile terminals and multiple intelligent base stations under the condition of minimum combined overhead of time delay and energy consumption.
Background
In recent years, with the increasing increase of user data, the large-scale access of internet of things devices and the diversification of services, the rapid increase of data traffic and the rapid expansion of data scale in the current wireless network are caused. Meanwhile, the functions of the mobile terminal are gradually enhanced, and the functions of the mobile terminal are not limited to the communication field, but become powerful carriers for people to move entertainment, work, read and calculate. Therefore, various highly complex terminal-dependent programs provided by third parties are beginning to be used in mobile computing in large numbers. However, the computing power of the terminal is limited by the volume, and the current battery technology has not been developed in a breakthrough manner, which brings great pressure to the field of mobile computing. In order to increase the processing speed of mobile communication services, reduce the delay of data transmission, and improve the user experience, the mobile communication industry is discussing that a mobile edge computing server is provided at the edge of a wireless access network (e.g., in a base station), and the mobile edge computing server can provide computing power and storage power for users accessing the wireless network.
A plurality of mobile edge computing tasks and data required by the tasks can be deployed on one mobile edge computing server according to needs, and each mobile edge computing task combination is utilized to complete a specific function and provide corresponding services. For example, tasks such as a "feature point identification task", a "distance calculation task", a "deviation angle calculation task", a "model drawing task", a "model tracking task", and a "model redrawing task" may be deployed in the mobile edge calculation server, and the calculated data required by the high-frequency task is cached in the storage device of the server, and when the mobile terminal user executes an augmented reality application, the tasks such as the "feature point identification task", the "model drawing task", and the "model tracking task" that have been deployed by the base station are used to implement the tasks; when a mobile terminal user executes an application of the internet of vehicles, the application is executed by utilizing a distance calculation task and an offset angle calculation task in the base station server. Compared with the traditional method, the method has the advantages that the original data acquired by the user equipment are uploaded to the remote server, the remote server carries out calculation based on the original data, identifies the target needing to be enhanced, acquires the information needing to be enhanced for the target according to the identified target, and finally transmits the information back to the user equipment, so that the identification speed of the target is increased, and the speed of pushing the enhanced information to the user terminal is increased.
In the process of implementing the invention, the inventor finds that the following problems exist in the prior art: in the prior art, mobile edge computing applications can be divided according to task types, and mobile edge computing task units capable of completing various mobile edge computing applications are deployed on an intelligent base station, when the number of users accessing a certain intelligent base station is large, the computing capacity of a mobile edge computing server is insufficient, and the users need to queue for service, so that the problems of user connection interruption or response overtime and the like are caused. Therefore, the mobile edge computing application used by the terminal may be limited by the service capabilities of the base station itself.
Disclosure of Invention
Technical problem to be solved
The technical problem to be solved by the invention is as follows: the embodiment of the invention provides a mobile edge computing application control method and resource scheduling, wherein a central server is used for distributing mobile edge computing application requests to complete load sharing operation and ensure that the system energy consumption and time delay cost are minimum and the profit is maximum, the problems of user connection interruption, response timeout and the like are avoided, the processing efficiency of services of mobile edge computing application is improved, and meanwhile, the collaborative optimization computing time delay and the terminal energy consumption are ensured.
(II) technical scheme
In order to solve the above technical problem, in a first aspect, the present invention provides a control method, where the mobile edge computing application control method includes:
the central base station, the base station has deployed the central control server, provide and schedule and data buffer memory function, the server includes:
the cache unit is used for caching and calculating required data according to task types, storing data with large relative quantity, caching high-frequency task data and reducing data access to a core network;
the control unit is used for controlling the target server selection of the mobile edge computing application according to the use condition of each current intelligent base station;
a receiving unit: the system comprises a server, a server and a server, wherein the server is used for receiving a service request from a mobile terminal and determining whether data corresponding to the service is cached or not;
a sending unit, configured to send a task data request that is not cached to a core network according to a determination result of the receiving unit if the cache unit does not include corresponding data in the service request; and if the cache unit contains the corresponding data in the service request, sending task data and a task request to a target server.
A serving base station, the base station having deployed therein an application computing server providing task computing functionality, the server comprising:
a receiving unit, configured to receive data required for task calculation from a central server and data required for task calculation from a core network;
the computing unit is used for computing a mobile edge computing task according to the task request and the task data;
and the sending unit is used for sending the service request calculation result to the mobile terminal.
In a second aspect, the present invention provides a resource scheduling method, including the steps of:
the method comprises the following steps: when detecting that a calculation task exists in the mobile terminal, sending a calculation request to a central base station;
step two: in each scheduling time slot, the central base station server calculates a decision matrix under the condition of minimum current cost according to the service conditions of all current intelligent base stations;
step three: if the decision matrix is inconsistent with the previous time slot, updating the scheduling strategy of the time slot; repeating the second step until the decision matrix is not changed any more, and the system reaches Nash equilibrium;
step four: determining whether a task needs to be executed on the intelligent base station according to a scheduling strategy, and determining a target intelligent base station of the calculation task;
step five: when the cache unit of the central base station has the calculation data required by the task contained in the request, the calculation data required by the task is sent to the task target intelligent base station; when the cache unit of the central base station lacks the calculation data required by the task contained in the request, the required task data requirement is sent to the network side;
step six: the target base station receives the required task data returned by the central base station and the network side;
step seven: and executing the calculation task according to the received required task data.
In some embodiments of the present invention, the application that includes K target servers and runs on all M terminals includes F types of tasks, where F = ∑ a mapping unit<c 1 ,d 1 >,<c 2 ,d 2 >,…,<c f ,d f >, where c, d represent the computation and data, respectively, required for a certain task. With X = { X 1 ,x 2 ,…,x f Indicates whether data required by a certain computation task is buffered in the central intelligent base station, x is a binary quantity and is represented by 0,1, 0 indicates that the data is not buffered, and 1 indicates that the data is buffered.
In some embodiments of the invention, the second step comprises: in each scheduling time slot, the central base station server calculates a decision matrix under the condition of minimum current cost according to the service condition of each current intelligent base station.
A game of gambling where the player has a limited set of actions, only a limited number of times being comparable, each of the limited broad forms of gambling with perfect information having a nash balance of pure strategy, can achieve a mutually satisfactory solution when the mobile device user is in equilibrium. The central base station counts the service conditions of the current intelligent service base stationsAnd calculating the scheme with the minimum current cost, and randomly selecting a terminal migration strategy for modification in each time slot until the decision matrix is not changed any more. According to y ij The values of the decision matrix determine which tasks in the mobile terminal application should be migrated and the proportion of the migrated tasks.
In some embodiments of the invention, step five comprises: judging whether data required by application requested by a terminal is cached in the intelligent base station or not, and sending the calculated data required by the task to the task target intelligent base station when the calculated data required by the task contained in the request exists in a cache unit of the central base station; and when the cache unit of the central base station lacks the calculation data required by the task contained in the request, sending a required task data requirement to the network side.
By usingIndicating the sniff delay due to the use of the central base station scheduling.
If d is j And after the cache is obtained, the central base station directly sends the calculation data to the target calculation base station, so that transmission delay is brought.
By usingRepresents the time delay caused by the buffered data required for the task contained in the application of the terminal i:
if d is j If the data is not cached, the data is sent to the core network to request for acquiring the data required by the task, so that extra time delay is brought.
By usingRepresents the additional delay due to the fact that the data required for the tasks contained in the application of terminal i are not buffered:
wherein,
p ij representing the request proportion of the terminal i to the task j;
λ i representing the request rate of the terminal i, the present invention considers the request from each terminal to be a poisson process;
representing the storage of calculation data d j Unit time delay between the core network and the central intelligent base station;
representing the presence of calculation data d j Unit time delay between the core network and the target intelligent base station;
representing unit time delay between the central intelligent base station and the terminal i;
representing unit time delay between the terminal i and the target intelligent base station;
h (req) and h (res) Respectively representing the lengths of the request message and the response message;
in some embodiments of the invention, the transmission delay of the terminal i comprises the transmission delay of the terminal to the central base stationTransmission delay from target base station to terminalSniffing latency:
wherein:
in some embodiments of the invention, the terminal computation time delay comprises a task terminal computation timeTask target base station computation timeTask sniffing computation time
Wherein:
known from queuing theory:
wherein:
where mu is the service rate of the target serving base station
In general:the calculation time of sniffing inquiry approaches to 0 and is ignored
In some embodiments of the present invention, the calculated migration delay of terminal i:
(III) advantageous effects
Compared with the prior art that the mobile edge computing application is divided according to the task types, and the mobile edge computing task units capable of completing various mobile edge computing applications are deployed on the intelligent base stations, when the number of users visiting a certain intelligent base station is large, the computing capacity of the mobile edge computing server is insufficient, the users need to queue for service, and therefore user connection interruption or response overtime occurs. The whole calculation migration process is regarded as a game with a limited action set, and when the users of the mobile devices are in a balanced state, a mutually satisfactory solution can be realized.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
Fig. 1 is a structural diagram of a central base station deployment scheme in an MEC provided in the present invention.
Fig. 2 is a structural diagram of a serving base station deployment scheme in the MEC provided by the present invention.
Fig. 3 is a flowchart of multi-server resource scheduling in the MEC provided by the present invention.
Fig. 4 is a block diagram of an implementation of multi-server resource scheduling in an MEC provided by the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention, but are not intended to limit the scope of the invention.
As shown in fig. 1, a structure diagram of a central base station deployment scenario according to the present invention is shown. The central base station comprises four units: a receiving unit 101, configured to receive a service request from a mobile terminal; a control unit 102, configured to determine whether a task scheduling policy of a certain terminal needs to be modified according to a received service request and a current use condition of each target server, and determine whether data corresponding to the service is cached; the cache unit 103 is used for caching the required data according to the task type, caching the high-frequency task data and reducing the access to the core network data; a sending unit 104, configured to send a task data request that is not cached to a core network when the control unit determines that the cache unit does not include corresponding data in the service request; and if the data is cached, sending the data required by calculation to the target server.
As shown in fig. 2, a structure diagram of a serving base station deployment scenario according to the present invention. The base station comprises three units: a receiving unit 201, configured to receive buffered data from a central base station and receive non-buffered data from a core network; a calculating unit 202, configured to calculate a calculation task migrated from the mobile terminal; a sending unit 203, configured to send the service request calculation result to the mobile terminal.
Fig. 3 is a flowchart of multi-server resource scheduling in an MEC according to the present invention. The invention provides a multi-server resource scheduling method, which comprises the following steps:
step 301: when a computing task is detected in the mobile terminal, a computing migration request is sent to the central intelligent base station;
as an example, an application running on a mobile terminal needs to be divided into two divisions F = &<c 1 ,d 1 >,<c 2 ,d 2 >,…,<c f ,d f >, sets of data, for the jth set of data, d j And c j Respectively representing the calculation amount and the data amount required by the j-th calculation task.
Step 302: in each scheduling time slot, the central base station server calculates a decision matrix under the condition of minimum current cost according to the service condition of each current intelligent base station.
Step 303: if the cost of the time slot is superior to that of the previous time slot, selecting a mobile terminal to change the decision matrix of the mobile terminal; and the system reaches Nash equilibrium until the decision matrix is not changed any more, and the decision matrix generated at the moment is the final decision matrix.
The resource scheduling of the whole system can be regarded as game games with limited action sets of players, only limited times are equivalent, the central base station can schedule in sequence according to the use condition of the current system, each game has Nash equilibrium of pure strategies, and when the users of the mobile devices are in a balanced state, a mutually satisfactory solution can be realized. And the central base station calculates the scheme with the minimum current cost according to the service conditions of all the current intelligent service base stations, and randomly selects a migration strategy of a terminal in each time slot to modify until the decision matrix is not changed any more.
Step 304, decision matrix no longer changes according to y ij The values of the decision matrix determine which tasks in the mobile terminal application should be migrated and the proportion of the migrated tasks.
And 305, determining a target server according to the decision matrix, and performing calculation migration.
As an embodiment, the central intelligent base station judges whether a data request needs to be sent to the network side according to whether the cache unit contains data required in the request. With vector X = { X 1 ,x 2 ,…,x f Denotes whether the task data is buffered, where x j Whether the intelligent base station caches the data required by the j-th calculation task is represented by a binary quantity which is represented by 0 and 1, wherein 0 represents that the data is not cached, and 1 represents that the data is cached.
In the process of calculating and transferring, transmission delay mainly exists between a mobile terminal and a central intelligent base station, and between a target intelligent base station and the mobile terminal, and sniffing transmission delay brought by using central base station scheduling comprises the following steps: if the calculated data is not cached, the time delay exists between the central intelligent base station and the server storing the data required by calculation, between the server storing the data required by calculation and the target intelligent base station, and if the data is cached, the time delay exists between the central intelligent base station and the target intelligent base station.
Here, by h (req) And h (res) Respectively representing the lengths of the request message and the response message, the invention considers that the request from each terminal is a Poisson process, lambda i Request rate on behalf of terminal i; p is a radical of ij Representing the request proportion of the terminal i to the task j and satisfyingThe ratio may be different for different terminals;representing the presence of calculation data d j Unit time delay between the core network and the central intelligent base station;representing the storage of calculation data d j Unit time delay between the core network and the target intelligent base station;representing unit time delay between the central intelligent base station and the terminal i;representing unit time delay between the terminal i and the target intelligent base station;
the transmission delay for terminal i can be expressed as follows:
by usingIndicating the sniff delay due to the use of the central base station scheduling.
If d is j And after the cache is obtained, the central base station directly sends the calculation data to the target calculation base station, so that transmission delay is brought. By usingRepresents the time delay caused by the buffered data required for the task contained in the application of the terminal i:
if d is j If the data is not cached, the data is sent to the core network to request to acquire the data required by the task, and extra time delay is brought. By usingRepresents the additional delay due to the fact that the data required for the tasks contained in the application of terminal i are not cached:
transmission delay from the terminal to the central base station:
transmission delay from the target base station to the terminal:
in the calculation and migration process, the calculation time comprises the operation time of the task terminalTask target base station computing timeTask sniffing computation timeThe intelligent base station uses a calculation migration decision matrix Y = (Y) ij ) M×F To manage the computation migration between the mobile terminal and the intelligent base station, where y ij (1. Ltoreq. I.ltoreq.M, 1. Ltoreq. J.ltoreq.F) in a representative terminal iAnd the task j is migrated to the proportion calculated by the intelligent base station. The service rate of the intelligent base station is represented by mu and theta i The service rate of the terminal i is represented, and the unit calculation time delay of the intelligent base station can be obtained according to the queuing theoryUnit calculation time delay of terminalWherein y is ij p ij λ i c j Representing the request rate of task j in terminal i to be transferred to intelligent base station to execute, (1-y) ij )p ij λ i c j Representing the request rate of local execution of task j in terminal i.
In general:the sniff poll computation time approaches 0 and is ignored.
The calculated delay for terminal i can be expressed as follows:
wherein:
known from the queuing theory:
for terminal i, the total delayTherefore, the total time delay of the system is equal to the sum of the time delays of all base stations in the coverage area of the base station, and can be expressed as:
as an embodiment, the calculation task is not migrated, and the unit calculation time delay of the terminal Time required for local execution:
as an embodiment, the MEC server generates the delayed revenue based on the delay gain
ΔZ＝Z loc -Z (14)
As an example, when Δ Z takes a maximum value and the maximum value of Δ Z of the next slot is unchanged, the system reaches a balance that satisfies the maximum satisfaction of each user. According to y ij The values of the decision matrix determine which tasks in the mobile terminal application should be migrated and the proportion of the migrated tasks.
As shown in FIG. 4, a block diagram of a multi-server resource scheduling implementation in accordance with the present invention is shown. The mobile terminal comprises three modules, namely a sending module: when the terminal has a calculation task, initiating a calculation migration request to a central intelligent base station; a receiving module: receiving a calculation migration decision matrix returned by the central intelligent base station and a calculation result of the target intelligent base station; a calculation module: and executing the local calculation task according to the decision matrix. The intelligent base station can be divided into a central scheduling control base station and a task target base station, a mobile edge computing server and a database are arranged in the central scheduling control base station, the task scheduling control base station controls task scheduling and stores high-frequency data, when data required by computing are not stored in the central intelligent base station, a request is sent to a server in a core network to obtain the data, and the task target base station is provided with the mobile edge computing server to perform task computing. The core network stores the data needed by the central base station for the stored calculation.
In summary, the method for controlling and scheduling resources of multi-server mobile edge computing provided by the invention creatively adopts the combination of the central base station and the service base station, the central base station is responsible for task scheduling and data caching, and the service base station is responsible for computing, so that the infrastructure and computing resources of the network are fully utilized, and the processing time of tasks is greatly shortened; moreover, according to different tasks applied by a user, the capability of task-based migration and proportion migration is provided for the terminal, so that the migration is more flexible; the whole calculation migration process is regarded as a game with a limited action set, and when the users of the mobile devices are in a balanced state, a mutually satisfactory solution can be realized. Meanwhile, after the central intelligent base station receives the task request sent by the terminal, whether data required by the task are cached in the central intelligent base station is judged, and only un-cached calculation data need to be obtained from the core network, so that the flexibility of the intelligent base station is improved, and the service processing efficiency is further improved; finally, the whole multi-server mobile edge computing control and resource scheduling method is compact and easy to control, and has wide and important popularization significance.
The above embodiments are only for illustrating the invention and are not to be construed as limiting the invention, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention, therefore, all equivalent technical solutions also belong to the scope of the invention, and the scope of the invention is defined by the claims.
Claims (8)
1. A control and resource scheduling method based on mobile edge computing comprising multiple servers is characterized by comprising the following steps:
step 301: when a computing task is detected in the mobile terminal, a computing migration request is sent to the central intelligent base station;
step 302: in each scheduling time slot, the central base station server calculates a decision matrix under the condition of minimum current cost according to the service conditions of all the current intelligent base stations.
Step 303: if the cost of the time slot is superior to that of the previous time slot, selecting a mobile terminal to change the decision matrix of the mobile terminal; and the system reaches Nash equilibrium until the decision matrix is not changed any more, and the generated decision matrix is the final decision matrix.
Step 304, the decision matrix is no longer changed according to y ij The values of the decision matrix determine which tasks in the mobile terminal application should be migrated and the proportion of the migrated tasks.
And 305, determining a target server according to the decision matrix, and performing calculation migration.
2. The method of claim 1, wherein the computation time for the computation migration of the application running on a terminal includes a time delay caused during transmission, a time delay for running computation, and a sniffing time delay for allocating resources with the over-center smart base station.
If the calculation data is not cached, the transmission delay also exists between the central intelligent base station and a server storing the data required by calculation, and between the server storing the data required by calculation and the target service base station. Here, by h (req) And h (res) Respectively representing the lengths of the request message and the response message, the invention considers the request from each terminal to be a Poisson process, lambda i Request rate on behalf of terminal i; p is a radical of ij Represents the request proportion of the terminal i to the task j and satisfiesThe ratio may be different for different terminals;representStoring calculation data d j Unit time delay between the core network and the central intelligent base station;representing the presence of calculation data d j Unit time delay between the core network and the target intelligent base station;representing unit time delay between the central intelligent base station and the terminal i;representing unit time delay between a terminal i and a target intelligent base station; the transmission delay for terminal i can be expressed as follows:
the calculation time comprises the operation time of the task terminalTask target base station computation timeTask sniffing computation timeThe intelligent base station uses a calculation migration decision matrix Y = (Y) ij ) M×F To manage the computation migration between the mobile terminal and the intelligent base station, where y ij And (i is more than or equal to 1 and less than or equal to M, and j is more than or equal to 1 and less than or equal to F) represents the calculated proportion of the task j in the terminal i migrated to the intelligent base station. The service rate of the intelligent base station is represented by mu and theta i Indicating the service rate of terminal iThe calculated delay for terminal i can be expressed as follows:
in general:the sniff poll computation time approaches 0 and is ignored.
The calculated delay for terminal i can be expressed as follows:
3. the method of claim 2, wherein the total delay time Therefore, the total time delay of the system is equal to the sum of the time delays of all base stations within the coverage area of the base station, and can be expressed as:
4. the method of claim 3, wherein the calculation task is not migrated, and the unit calculation delay of the terminal is calculatedTime required for local execution:
5. method according to claims 1 to 4, characterized in that said central base station server is based on saidTime delay gain Δ T = T loc and-T, dynamically adjusting the terminal decision matrix if the time delay gain is larger than the last time slot until the gain is not changed any more, and outputting a final decision matrix to maximize the satisfaction of each user.
6. A central base station deployment scenario, wherein the base station comprises:
a receiving unit, configured to receive a service request from a mobile terminal;
the control unit is used for determining whether a task scheduling strategy of a certain terminal needs to be modified or not according to the received service request and the use condition of each current target server, and determining whether data corresponding to the service is cached or not;
the cache unit is used for caching and calculating the required data according to the task type, caching the high-frequency task data and reducing the access to the core network data;
and the sending unit is used for sending the task data request which is not cached to a core network or sending the data required by calculation to a target server according to the fact that whether the cache unit contains the corresponding data in the service request or not is judged by the control unit.
7. A serving base station deployment scenario, wherein the base station comprises:
the receiving unit is used for receiving the cache data from the central base station and receiving the non-cache data from the core network;
the computing unit is used for computing the computing task transferred from the mobile terminal;
and the sending unit is used for sending the service request calculation result to the mobile terminal.
8. The method according to any one of claims 1 to 7, wherein the mobile terminal performs the calculation migration operation according to the stable migration decision message, so as to maximize all user satisfaction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711011580.4A CN107734558A (en) | 2017-10-26 | 2017-10-26 | A kind of control of mobile edge calculations and resource regulating method based on multiserver |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711011580.4A CN107734558A (en) | 2017-10-26 | 2017-10-26 | A kind of control of mobile edge calculations and resource regulating method based on multiserver |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107734558A true CN107734558A (en) | 2018-02-23 |
Family
ID=61213629
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711011580.4A Pending CN107734558A (en) | 2017-10-26 | 2017-10-26 | A kind of control of mobile edge calculations and resource regulating method based on multiserver |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107734558A (en) |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509276A (en) * | 2018-03-30 | 2018-09-07 | 南京工业大学 | Video task dynamic migration method in edge computing environment |
| CN108632860A (en) * | 2018-04-17 | 2018-10-09 | 浙江工业大学 | A kind of mobile edge calculations rate maximization approach based on deeply study |
| CN108632861A (en) * | 2018-04-17 | 2018-10-09 | 浙江工业大学 | A kind of mobile edge calculations shunting decision-making technique based on deeply study |
| CN108632862A (en) * | 2018-04-17 | 2018-10-09 | 浙江工业大学 | A kind of mobile edge calculations shunting decision-making technique based on depth deterministic policy gradient method |
| CN108667850A (en) * | 2018-05-21 | 2018-10-16 | 济南浪潮高新科技投资发展有限公司 | A kind of artificial intelligence service system and its method for realizing artificial intelligence service |
| CN108738046A (en) * | 2018-04-17 | 2018-11-02 | 浙江工业大学 | A kind of mobile edge calculations rate maximization approach based on semi-supervised learning |
| CN108738045A (en) * | 2018-04-17 | 2018-11-02 | 浙江工业大学 | A kind of mobile edge calculations rate maximization approach based on depth deterministic policy gradient method |
| CN108901046A (en) * | 2018-06-14 | 2018-11-27 | 北京大学 | Cotasking unloading algorithm and system design scheme towards mobile edge calculations |
| CN108924198A (en) * | 2018-06-21 | 2018-11-30 | 中国联合网络通信集团有限公司 | A kind of data dispatching method based on edge calculations, apparatus and system |
| CN108984307A (en) * | 2018-07-19 | 2018-12-11 | 中国联合网络通信集团有限公司 | Calculating task moving method and calculating task transference apparatus |
| CN109002357A (en) * | 2018-06-07 | 2018-12-14 | 阿里巴巴集团控股有限公司 | Resource allocation methods, device and Internet of things system |
| CN109120583A (en) * | 2018-06-13 | 2019-01-01 | 深圳市海派通讯科技有限公司 | A method of the buffer encrypted data based on action boundary operation |
| CN109167671A (en) * | 2018-07-11 | 2019-01-08 | 国网信通亿力科技有限责任公司 | A kind of adapted communication system equally loaded dispatching algorithm towards quantum key distribution business |
| CN109391681A (en) * | 2018-09-14 | 2019-02-26 | 重庆邮电大学 | V2X mobility prediction based on MEC unloads scheme with content caching |
| CN109587735A (en) * | 2018-11-12 | 2019-04-05 | 电子科技大学 | A kind of cooperation with service caching method for measuring mobile edge calculations based on time delay |
| CN109709985A (en) * | 2019-01-21 | 2019-05-03 | 中国联合网络通信集团有限公司 | A kind of unmanned plane task optimization method, apparatus and system |
| CN109862592A (en) * | 2018-12-06 | 2019-06-07 | 北京邮电大学 | Resource management and dispatching method under a kind of mobile edge calculations environment based on multi-base station cooperative |
| CN110022300A (en) * | 2019-03-06 | 2019-07-16 | 百度在线网络技术（北京）有限公司 | Edge calculations implementation method, equipment, system and storage medium |
| CN110248206A (en) * | 2019-07-29 | 2019-09-17 | 北京邮电大学 | A kind of resource allocation methods, device and electronic equipment for edge network system |
| CN110324853A (en) * | 2019-07-12 | 2019-10-11 | 中国科学院计算技术研究所 | A kind of anti-Caton mobile communications network switching method and corresponding network framework |
| CN111131272A (en) * | 2019-12-27 | 2020-05-08 | 广州酷狗计算机科技有限公司 | Scheduling method, device and system of stream server, computing equipment and storage medium |
| CN111611069A (en) * | 2020-04-01 | 2020-09-01 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Multi-type task migration method among multiple data centers |
| CN111857976A (en) * | 2020-08-07 | 2020-10-30 | 华侨大学 | Multi-objective optimization calculation migration method based on decomposition |
| CN112182970A (en) * | 2020-09-30 | 2021-01-05 | 天津大学 | Edge calculation deployment method for minimizing delay of nondestructive testing |
| CN112367605A (en) * | 2019-07-25 | 2021-02-12 | 中兴通讯股份有限公司 | Application relocation method and device based on edge calculation |
| CN112887347A (en) * | 2019-11-29 | 2021-06-01 | 中国信息通信研究院 | Dynamic migration method and device for edge calculation in industrial internet |
| CN113032576A (en) * | 2019-12-25 | 2021-06-25 | 杭州海康威视数字技术股份有限公司 | Knowledge graph management method, device, equipment and storage medium |
| CN113692052A (en) * | 2021-08-05 | 2021-11-23 | 广东技术师范大学 | Network edge machine learning training method |
| CN113709817A (en) * | 2021-08-13 | 2021-11-26 | 北京信息科技大学 | Task unloading and resource scheduling method and device under multi-base-station multi-server scene |
| CN114125936A (en) * | 2021-11-29 | 2022-03-01 | 中国联合网络通信集团有限公司 | Resource scheduling method, device and storage medium |
| CN114398162A (en) * | 2021-12-30 | 2022-04-26 | 洛阳隆盛科技有限责任公司 | Edge computing task cooperative scheduling method |
| CN114449486A (en) * | 2021-12-23 | 2022-05-06 | 之江实验室 | Method and device for roaming edge computing service |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106100907A (en) * | 2016-08-15 | 2016-11-09 | 北京邮电大学 | A kind of MEC server selection algorithm based on fairness |
| CN106534333A (en) * | 2016-11-30 | 2017-03-22 | 北京邮电大学 | Bidirectional selection computing unloading method based on MEC and MCC |
| CN106900011A (en) * | 2017-02-28 | 2017-06-27 | 重庆邮电大学 | Task discharging method between a kind of cellular basestation based on MEC |
-
2017
- 2017-10-26 CN CN201711011580.4A patent/CN107734558A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106100907A (en) * | 2016-08-15 | 2016-11-09 | 北京邮电大学 | A kind of MEC server selection algorithm based on fairness |
| CN106534333A (en) * | 2016-11-30 | 2017-03-22 | 北京邮电大学 | Bidirectional selection computing unloading method based on MEC and MCC |
| CN106900011A (en) * | 2017-02-28 | 2017-06-27 | 重庆邮电大学 | Task discharging method between a kind of cellular basestation based on MEC |
Non-Patent Citations (1)
| Title |
|---|
| 张建敏 等: "移动边缘计算技术及其本地分流方案", 《电信科学》 * |
Cited By (52)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509276A (en) * | 2018-03-30 | 2018-09-07 | 南京工业大学 | Video task dynamic migration method in edge computing environment |
| CN108509276B (en) * | 2018-03-30 | 2021-11-30 | 南京工业大学 | Video task dynamic migration method in edge computing environment |
| CN108738046A (en) * | 2018-04-17 | 2018-11-02 | 浙江工业大学 | A kind of mobile edge calculations rate maximization approach based on semi-supervised learning |
| CN108632862B (en) * | 2018-04-17 | 2021-06-18 | 浙江工业大学 | Moving edge calculation shunting decision method based on depth certainty strategy gradient |
| CN108738046B (en) * | 2018-04-17 | 2021-04-06 | 浙江工业大学 | Mobile edge calculation rate maximization method based on semi-supervised learning |
| CN108738045B (en) * | 2018-04-17 | 2021-04-06 | 浙江工业大学 | Moving edge calculation rate maximization method based on depth certainty strategy gradient |
| CN108738045A (en) * | 2018-04-17 | 2018-11-02 | 浙江工业大学 | A kind of mobile edge calculations rate maximization approach based on depth deterministic policy gradient method |
| CN108632861B (en) * | 2018-04-17 | 2021-06-18 | 浙江工业大学 | Mobile edge calculation shunting decision method based on deep reinforcement learning |
| CN108632862A (en) * | 2018-04-17 | 2018-10-09 | 浙江工业大学 | A kind of mobile edge calculations shunting decision-making technique based on depth deterministic policy gradient method |
| CN108632860B (en) * | 2018-04-17 | 2021-06-18 | 浙江工业大学 | Mobile edge calculation rate maximization method based on deep reinforcement learning |
| CN108632860A (en) * | 2018-04-17 | 2018-10-09 | 浙江工业大学 | A kind of mobile edge calculations rate maximization approach based on deeply study |
| CN108632861A (en) * | 2018-04-17 | 2018-10-09 | 浙江工业大学 | A kind of mobile edge calculations shunting decision-making technique based on deeply study |
| CN108667850A (en) * | 2018-05-21 | 2018-10-16 | 济南浪潮高新科技投资发展有限公司 | A kind of artificial intelligence service system and its method for realizing artificial intelligence service |
| CN108667850B (en) * | 2018-05-21 | 2020-10-27 | 浪潮集团有限公司 | Artificial intelligence service system and method for realizing artificial intelligence service |
| CN109002357A (en) * | 2018-06-07 | 2018-12-14 | 阿里巴巴集团控股有限公司 | Resource allocation methods, device and Internet of things system |
| CN109002357B (en) * | 2018-06-07 | 2021-11-23 | 创新先进技术有限公司 | Resource allocation method and device and Internet of things system |
| CN109120583A (en) * | 2018-06-13 | 2019-01-01 | 深圳市海派通讯科技有限公司 | A method of the buffer encrypted data based on action boundary operation |
| CN108901046A (en) * | 2018-06-14 | 2018-11-27 | 北京大学 | Cotasking unloading algorithm and system design scheme towards mobile edge calculations |
| CN108924198A (en) * | 2018-06-21 | 2018-11-30 | 中国联合网络通信集团有限公司 | A kind of data dispatching method based on edge calculations, apparatus and system |
| CN109167671A (en) * | 2018-07-11 | 2019-01-08 | 国网信通亿力科技有限责任公司 | A kind of adapted communication system equally loaded dispatching algorithm towards quantum key distribution business |
| CN109167671B (en) * | 2018-07-11 | 2023-04-28 | 国网信通亿力科技有限责任公司 | Quantum key distribution service-oriented balanced load scheduling method for distribution communication system |
| CN108984307A (en) * | 2018-07-19 | 2018-12-11 | 中国联合网络通信集团有限公司 | Calculating task moving method and calculating task transference apparatus |
| CN108984307B (en) * | 2018-07-19 | 2020-11-10 | 中国联合网络通信集团有限公司 | Computing task migration method and computing task migration device |
| CN109391681A (en) * | 2018-09-14 | 2019-02-26 | 重庆邮电大学 | V2X mobility prediction based on MEC unloads scheme with content caching |
| CN109391681B (en) * | 2018-09-14 | 2021-07-16 | 重庆邮电大学 | MEC-based V2X mobility prediction and content caching offloading scheme |
| CN109587735A (en) * | 2018-11-12 | 2019-04-05 | 电子科技大学 | A kind of cooperation with service caching method for measuring mobile edge calculations based on time delay |
| CN109587735B (en) * | 2018-11-12 | 2021-07-23 | 电子科技大学 | Cooperative service caching method based on time delay measurement mobile edge calculation |
| CN109862592B (en) * | 2018-12-06 | 2020-11-10 | 北京邮电大学 | Resource management and scheduling method in mobile edge computing environment based on multi-base-station cooperation |
| CN109862592A (en) * | 2018-12-06 | 2019-06-07 | 北京邮电大学 | Resource management and dispatching method under a kind of mobile edge calculations environment based on multi-base station cooperative |
| CN109709985A (en) * | 2019-01-21 | 2019-05-03 | 中国联合网络通信集团有限公司 | A kind of unmanned plane task optimization method, apparatus and system |
| CN110022300A (en) * | 2019-03-06 | 2019-07-16 | 百度在线网络技术（北京）有限公司 | Edge calculations implementation method, equipment, system and storage medium |
| CN110324853A (en) * | 2019-07-12 | 2019-10-11 | 中国科学院计算技术研究所 | A kind of anti-Caton mobile communications network switching method and corresponding network framework |
| CN112367605A (en) * | 2019-07-25 | 2021-02-12 | 中兴通讯股份有限公司 | Application relocation method and device based on edge calculation |
| CN110248206A (en) * | 2019-07-29 | 2019-09-17 | 北京邮电大学 | A kind of resource allocation methods, device and electronic equipment for edge network system |
| CN112887347A (en) * | 2019-11-29 | 2021-06-01 | 中国信息通信研究院 | Dynamic migration method and device for edge calculation in industrial internet |
| CN112887347B (en) * | 2019-11-29 | 2022-10-28 | 中国信息通信研究院 | Dynamic migration method and device for edge calculation in industrial internet |
| CN113032576B (en) * | 2019-12-25 | 2022-06-03 | 杭州海康威视数字技术股份有限公司 | Knowledge graph management method, device, equipment and storage medium |
| CN113032576A (en) * | 2019-12-25 | 2021-06-25 | 杭州海康威视数字技术股份有限公司 | Knowledge graph management method, device, equipment and storage medium |
| CN111131272A (en) * | 2019-12-27 | 2020-05-08 | 广州酷狗计算机科技有限公司 | Scheduling method, device and system of stream server, computing equipment and storage medium |
| CN111611069A (en) * | 2020-04-01 | 2020-09-01 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Multi-type task migration method among multiple data centers |
| CN111611069B (en) * | 2020-04-01 | 2023-11-07 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | Multi-type task migration method among multiple data centers |
| CN111857976A (en) * | 2020-08-07 | 2020-10-30 | 华侨大学 | Multi-objective optimization calculation migration method based on decomposition |
| CN111857976B (en) * | 2020-08-07 | 2023-06-02 | 华侨大学 | Multi-objective optimization computing migration method based on decomposition |
| CN112182970A (en) * | 2020-09-30 | 2021-01-05 | 天津大学 | Edge calculation deployment method for minimizing delay of nondestructive testing |
| CN113692052A (en) * | 2021-08-05 | 2021-11-23 | 广东技术师范大学 | Network edge machine learning training method |
| CN113709817B (en) * | 2021-08-13 | 2023-06-06 | 北京信息科技大学 | Task unloading and resource scheduling method and device under multi-base-station multi-server scene |
| CN113709817A (en) * | 2021-08-13 | 2021-11-26 | 北京信息科技大学 | Task unloading and resource scheduling method and device under multi-base-station multi-server scene |
| CN114125936A (en) * | 2021-11-29 | 2022-03-01 | 中国联合网络通信集团有限公司 | Resource scheduling method, device and storage medium |
| CN114125936B (en) * | 2021-11-29 | 2023-09-05 | 中国联合网络通信集团有限公司 | Resource scheduling method, device and storage medium |
| CN114449486A (en) * | 2021-12-23 | 2022-05-06 | 之江实验室 | Method and device for roaming edge computing service |
| CN114449486B (en) * | 2021-12-23 | 2023-09-19 | 之江实验室 | Method and device for roaming edge computing service |
| CN114398162A (en) * | 2021-12-30 | 2022-04-26 | 洛阳隆盛科技有限责任公司 | Edge computing task cooperative scheduling method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107734558A (en) | A kind of control of mobile edge calculations and resource regulating method based on multiserver | |
| CN109862592B (en) | Resource management and scheduling method in mobile edge computing environment based on multi-base-station cooperation | |
| CN112995023B (en) | Multi-access edge computing network computing unloading system and computing unloading method thereof | |
| CN111586696B (en) | Resource allocation and unloading decision method based on multi-agent architecture reinforcement learning | |
| CN107846704A (en) | A kind of resource allocation and base station service arrangement method based on mobile edge calculations | |
| CN111124662B (en) | Fog calculation load balancing method and system | |
| CN113950066A (en) | Single server part calculation unloading method, system and equipment under mobile edge environment | |
| CN110933139A (en) | System and method for solving high concurrency of Web server | |
| CN102662764B (en) | A kind of dynamic cloud computational resource optimizing distribution method based on SMDP | |
| CN109474681A (en) | Resource allocation methods, system and the server system of mobile edge calculations server | |
| CN111400001A (en) | Online computing task unloading scheduling method facing edge computing environment | |
| CN111163178B (en) | Game theory-based service deployment and task unloading method in edge computing | |
| CN111552564A (en) | Task unloading and resource optimization method based on edge cache | |
| CN113993218A (en) | Multi-agent DRL-based cooperative unloading and resource allocation method under MEC architecture | |
| CN111988787B (en) | Task network access and service placement position selection method and system | |
| CN110069341A (en) | What binding function configured on demand has the dispatching method of dependence task in edge calculations | |
| CN112888005B (en) | MEC-oriented distributed service scheduling method | |
| Chen et al. | DDPG-based computation offloading and service caching in mobile edge computing | |
| CN116321307A (en) | Bidirectional cache placement method based on deep reinforcement learning in non-cellular network | |
| CN109327340B (en) | Mobile wireless network virtual network mapping method based on dynamic migration | |
| CN113329432B (en) | Edge service arrangement method and system based on multi-objective optimization | |
| Nguyen et al. | EdgePV: collaborative edge computing framework for task offloading | |
| CN113778675A (en) | Calculation task distribution system and method based on block chain network | |
| CN109600432B (en) | Dynamic access method for content center mobile edge network user | |
| CN116939044A (en) | Computing power route planning method and device based on block chain technology |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180223 |