CN116132443B - Management scheduling system and method for ubiquitous edge computing - Google Patents
Management scheduling system and method for ubiquitous edge computing Download PDFInfo
- Publication number
- CN116132443B CN116132443B CN202310037907.4A CN202310037907A CN116132443B CN 116132443 B CN116132443 B CN 116132443B CN 202310037907 A CN202310037907 A CN 202310037907A CN 116132443 B CN116132443 B CN 116132443B
- Authority
- CN
- China
- Prior art keywords
- edge
- subsystem
- edge node
- data
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000005259 measurement Methods 0.000 claims abstract description 56
- 238000004364 calculation method Methods 0.000 claims abstract description 3
- 238000007726 management method Methods 0.000 claims description 67
- 230000001133 acceleration Effects 0.000 claims description 7
- 238000013523 data management Methods 0.000 claims description 7
- 238000013500 data storage Methods 0.000 claims description 6
- 230000001419 dependent effect Effects 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000005012 migration Effects 0.000 abstract description 8
- 238000013508 migration Methods 0.000 abstract description 8
- 238000013461 design Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 7
- 230000001360 synchronised effect Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 239000012634 fragment Substances 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 238000009877 rendering Methods 0.000 description 2
- 102100037681 Protein FEV Human genes 0.000 description 1
- 101710198166 Protein FEV Proteins 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 239000012729 immediate-release (IR) formulation Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
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
-
- 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
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- 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
- H04W28/09—Management thereof
- H04W28/0925—Management thereof using policies
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
- H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Mobile Radio Communication Systems (AREA)
- Multi Processors (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a management scheduling system and a method for ubiquitous edge calculation, comprising a control layer and an edge layer, wherein the control layer comprises a central scheduling control system, and the edge layer automatically forms a mesh network by edge nodes to form an edge cloud; the central dispatching control system comprises a dispatching subsystem, a first application management subsystem, a data distribution subsystem, a first measurement subsystem, an edge node management subsystem and a mobile target management subsystem; the edge node comprises a data subsystem, a second measurement subsystem, a second application management subsystem and a control subsystem. Compared with the prior art, the method and the device have the advantages that application migration is not needed to be performed when the mobile target enters a new target area, and the edge node which is provided with the service at the latest can efficiently perform parallel downloading from a plurality of other edge nodes, so that the downloading efficiency is relatively higher, and the problem that smooth service is continuously provided for the mobile target in a mobile scene is solved.
Description
Technical Field
The invention relates to the technical field of the Internet of things, in particular to a management scheduling system and method for ubiquitous edge computing.
Background
In the edge computing field/the internet of things field/the internet of vehicles, how to provide low-delay, smooth and reliable service for a service object in a mobile scene is a troublesome problem. For the static service object, the problem of time delay can be solved by providing nearby services through fixed edge equipment. However, when the service object is in a high-speed motion process, which means that there is no fixed device for providing services to the service object, then the allocation of resources and data migration to the target service object in the process can be a troublesome problem. For example, when the vehicle is in a high speed motion, it is a difficulty how to provide a low latency smooth and reliable service for vehicular services and mobile application services used by passengers.
Currently, the commonly used schemes include the following two: first, when RSU (Road SideUnit) is in the vicinity of the vehicle, the RSU provides the vehicle service with the application service in the vicinity; second, when a vehicle accesses an MEC (Multi-access edge computing, multilateral access service), the MEC provides application services to the vehicle. The MEC herein may in some cases also be an RSU in the first scenario described above. Both schemes are based on providing application services on an edge layer like RSU or MEC near a service object, in an implementation, when the nearby edge layer has no services or data required by the service object, services and data synchronization are requested by the nearby edge cloud to the central cloud layer.
However, the conventional scheme is prone to the following problems when RSU or MEC handover occurs:
1) If the newly switched RSU or MEC does not have the service or data required by the service object at the nearby edge level, the RSU or MEC requests the data synchronization service at the cloud level of the center, and under the condition that the vehicle moves at a high speed, the time delay is uncontrollable, and it is very likely that the data synchronization is completed, the vehicle is not in the area, and when one vehicle is in the area, the RSU or MEC cannot provide the required service.
2) The service object requests a service from the central cloud, in which case the delay of the service is not controllable.
3) Under the current scheme, if the faster application deployment is to be provided, full synchronization of data is required, which has a high overhead for the whole network.
For the problems in the related art, no effective solution has been proposed at present.
Disclosure of Invention
Aiming at the problems in the related art, the invention provides a ubiquitous edge computing management scheduling system and a ubiquitous edge computing management scheduling method, so as to overcome the technical problems in the prior related art.
The invention solves the problem of how to provide smooth and reliable low-delay data service for a service target object under the condition that the service object moves at a high speed. Compared with the traditional scheme, the scheme needs to solve the problem of how to realize the noninductive migration of the application service for the target object when the target object moves at a high speed; this will involve very fast deployment of the application and accurate synchronization of the data. The invention is equally applicable when the moving object is stationary, meaning that there is no need to synchronize data to the surrounding base design.
For this purpose, the invention adopts the following specific technical scheme:
according to one aspect of the invention, a management scheduling system for ubiquitous edge computing is provided, the data management scheduling system comprises a control layer and an edge layer, the control layer comprises a central scheduling control system, and the edge layer automatically forms a mesh network by edge nodes to form an edge cloud;
the central dispatching control system comprises a dispatching subsystem, a first application management subsystem, a data distribution subsystem, a first measurement subsystem, an edge node management subsystem and a mobile target management subsystem;
the edge node comprises a data subsystem, a second measurement subsystem, a second application management subsystem and a control subsystem;
the scheduling subsystem is used for determining the data management and the application life cycle management of each edge node according to the monitoring information of each edge node;
the first application management subsystem is used for providing life cycle management of the global application and deciding to provide corresponding application services for the mobile target according to the subscription relation of the mobile target;
the data distribution subsystem is used for providing data slicing indexes, and each edge node decides a place for downloading data according to the slicing indexes;
the first measurement subsystem is used for measuring global speed measurement and helping a mobile target to select to access an edge service system; when a moving target initiates speed measurement, a speed measurement system of a control plane estimates a conforming target edge service system list according to the network time delay and the moving direction of the moving target, and a system which is recommended most preferentially is placed at the first place;
the edge node management subsystem is used for providing management and query services of each edge node, namely setting a grid range of the edge node service and an access mode of the edge node;
the mobile target management subsystem is used for managing the authentication data of the mobile target and the subscription relation;
the data subsystem is used for realizing the functions of data storage and distribution;
the second measurement subsystem is used for finding a moving target entering the covered area and establishing a communication link with the moving target; the method is also used for receiving the speed, acceleration, direction and position information of the moving target;
the second application management subsystem is used for managing the application life cycle according to the instruction of the control subsystem;
the control subsystem is used for predicting the advancing direction of the moving target according to the real-time motion data and map data of each moving target which are currently served and obtained by the second measurement subsystem, and estimating the probability of when the moving target leaves the grid range served by the current edge node and is expected to go to the target grid; but also to decide when to synchronize data to the edge nodes associated with which target grids and to trigger the target edge nodes to create the application services required by the moving target.
Further, adjacent edge nodes in the edge layer are automatically connected with each other, and a mesh network is formed.
Furthermore, all data of the whole data management scheduling system are stored in the edge nodes, and the central scheduling control system is also bound with global data.
Further, the data stored in the data subsystem includes data generated and dependent by an application running on the edge node, shadow data of a moving target of the grid covered by the edge node, and running data of the edge node itself.
Further, the data subsystem is further configured to synchronously distribute the data related to the moving object to the neighboring edge nodes when the moving object enters the coverage area of the other edge nodes.
Further, the application is responsible for providing services to specific moving targets, wherein the application comprises the following two types:
the first is to reside on the edge node for a long time, serving all moving targets within all grids hosted by the edge node;
the second is a personalized application actively applied to the system by the mobile target, which is used for serving the special service of the mobile target.
According to another aspect of the present invention, there is provided a management scheduling method of ubiquitous edge computing, for scheduling when a target terminal first enters a management scheduling system of ubiquitous edge computing to provide a service area, the scheduling method comprising the steps of:
the central dispatching control system informs the target terminal of the edge node which provides service for the grid according to the grid where the target terminal is located;
the target terminal receives the notification information and automatically establishes connection with an edge node in the notification information;
the target terminal reports measurement data to the connected edge node at regular time, and the second measurement subsystem is utilized for measurement;
the control subsystem in the edge node confirms that the mobile target enters for the first time according to the obtained measured value, and informs the data subsystem of synchronizing the ordering relation and the data on which the mobile target depends from the central dispatching control system;
after the data synchronization is completed, the control subsystem in the edge node inquires the application currently used by the target terminal to the central dispatching control system, if the application does not exist, the second application management subsystem is informed to carry out application creation, and after the application is available, the control subsystem is used for informing the target terminal of the application service address of the service target terminal;
and the target terminal requests the service from the edge area through the corresponding application service address, so that the target terminal obtains continuous service.
Further, the target terminal requests the service to the edge area through the application service address corresponding to the target terminal, and the method further comprises the following steps:
in the grid crossing area, when a plurality of edge nodes provide service at the same time, the target terminal selects a corresponding edge area according to the link quality of the edge nodes and requests the service from the edge area.
According to still another aspect of the present invention, there is provided a management scheduling method of ubiquitous edge calculation for scheduling when a target terminal is driven into an edge node Mb coverage grid by an edge node Ma coverage grid, the scheduling method comprising the steps of:
the edge node Ma judges the probability of the target terminal entering other edge node coverage grids according to the running state of the current target terminal, and sends a notification message to the corresponding edge node when the probability exceeds a preset value;
the second measurement subsystem Mb in the edge node discovers that the target terminal enters the grid covered by the target terminal, and indicates that the target terminal enters the edge node Mb area from the edge node Ma area;
when entering a grid crossing area, the edge node at the previous moment informs the target terminal of the edge node which provides service for the area, and the target terminal is connected with the edge node which provides service for the area;
the target terminal reports measurement data to the connected edge node at regular time, reports the last edge node providing service as an edge node Ma, and measures by utilizing a second measurement subsystem of the connected edge node;
the control subsystem in the edge node Mb judges the probability of the target terminal entering the edge node Mb area according to the driving direction and the speed of the target terminal, and when the predicted probability exceeds a preset threshold value, the control subsystem informs the data subsystem of synchronizing the ordering relation of the target terminal and the data relied by the moving target from the edge node Ma or the central dispatching control system in combination with the probability in the edge node Ma notification message;
after the data synchronization is completed, the control subsystem in the edge node Mb inquires the control subsystem in the edge node Ma or the central scheduling control system about the application currently used by the target terminal, informs the second application management subsystem of creating the application, and informs the target terminal of the completion of the application creation of the service target terminal in the edge node Mb area through the control subsystem after the application creation is completed;
and the target terminal requests the service from the edge area through the corresponding application service address, so that the target terminal obtains continuous service.
Further, the target terminal requests the service to the edge area through the application service address corresponding to the target terminal, and the method further comprises the following steps:
in the grid crossing area, when a plurality of edge nodes provide service at the same time, the target terminal selects a corresponding edge area according to the link quality of the edge nodes and requests the service from the edge area.
The beneficial effects of the invention are as follows:
1) According to the gridding design of the edge nodes, when the moving target enters the cross grids, the probability estimation is carried out, and the adjacent grids are triggered to carry out data synchronization, so that services can be provided for the moving target entering the new grids more quickly, smooth services can be provided for users, and the problem that smooth services are continuously provided for the moving target in a moving scene is solved. Compared with the prior art, the application migration is not required to be performed when the moving target enters a new target area.
2) According to the networking architecture provided by the invention, when the mobile target passes through one edge node, the edge node trigger records who the last edge node serves, so that when the mobile target needs to reach a new edge node, the system can know which edge nodes serve the target before, and can more efficiently know which edge nodes can obtain data when the data are synchronized, and the concurrent downloading can be performed quickly. Compared with the prior art, the system records the moving path of the target, so that the edge node which is provided with the service recently can efficiently perform parallel downloading from a plurality of other edge nodes, and the downloading efficiency is relatively higher.
3) The measurement services provided by the invention, and the design of probability estimation based thereon, provide support for when data migration and application migration are performed. Based on the probability estimation design, compared with the existing scheme, new edge service can be informed in advance to prepare data and application preparation in advance, and invalid preparation can be avoided.
4) The data storage structure of the system has no single node fault, and theoretically, each node which is recently served by each mobile target has more complete data than the last service node, and the service provided for the mobile target is not completely influenced by the fault of a certain node. Compared with the existing scheme, the storage scheme of the system has clear synchronous target objects, invalid data synchronization and obvious single-node faults.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a block diagram of a management scheduling system for ubiquitous edge computing, according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a management scheduling system for ubiquitous edge computing, according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of the first scenario in a management scheduling method for ubiquitous edge computing according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a second scenario in a management scheduling method for ubiquitous edge computing according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a predictive algorithm in a management scheduling method for ubiquitous edge computing according to an embodiment of the present invention;
fig. 6 is a schematic diagram of data synchronization in a management scheduling method of ubiquitous edge computing according to an embodiment of the present invention.
In the figure:
1. a central dispatch control system; 11. a scheduling subsystem; 12. a first application management subsystem; 13. a data distribution subsystem; 14. a first measurement subsystem; 15. an edge node management subsystem; 16. a moving target management subsystem; 2. edge nodes; 21. a data subsystem; 22. a second measurement subsystem; 23. a second application management subsystem; 24. and a control subsystem.
Detailed Description
For the purpose of further illustrating the various embodiments, the present invention provides the accompanying drawings, which are a part of the disclosure of the present invention, and which are mainly used to illustrate the embodiments and, together with the description, serve to explain the principles of the embodiments, and with reference to these descriptions, one skilled in the art will recognize other possible implementations and advantages of the present invention, wherein elements are not drawn to scale, and like reference numerals are generally used to designate like elements.
According to the embodiment of the invention, a system and a method for managing and scheduling ubiquitous edge computing are provided.
The invention will be further described with reference to the accompanying drawings and the specific embodiments, as shown in fig. 1-2, according to one embodiment of the invention, there is provided a ubiquitous edge computing management and scheduling system, where the data management and scheduling system includes a control layer and an edge layer, the control layer is responsible for a central scheduling and controlling system 1, the edge layer is formed into an edge cloud by automatically forming a mesh network by edge nodes 2, and adjacent edge nodes 2 in the edge layer are automatically connected with each other and form a mesh network.
The central dispatching control system 1 comprises a dispatching subsystem 11, a first application management subsystem 12, a data distribution subsystem 13, a first measurement subsystem 14, an edge node management subsystem 15 and a moving target management subsystem 16;
the edge node 2 comprises a data subsystem 21, a second measurement subsystem 22, a second application management subsystem 23 and a control subsystem 24;
the scheduling subsystem 11 is configured to determine data management and application lifecycle management of each edge node according to the monitoring information of each edge node;
the first application management subsystem 12 is configured to provide global application lifecycle management, and determine what application service to provide to the mobile object according to the subscription relationship of the mobile object;
the data distribution subsystem 13 is configured to provide a data slice index, and each edge node decides from where to download data according to the slice index. The data in the whole system is positioned at each edge node, and the central storage also binds global data to make disaster recovery backup;
the first measurement subsystem 14 is used for measuring global speed measurement and helping a mobile target to select to access an edge service system; when a moving target initiates speed measurement, a speed measurement system of a control plane estimates a conforming target edge service system list according to the network time delay and the moving direction of the moving target, and a system which is recommended most preferably is placed at the forefront;
the edge node management subsystem 15 is used to provide management and query services for the edge nodes. Setting the range of the grid served by the edge node and the access mode of the edge node;
the mobile target management subsystem 16 is used for being responsible for management of mobile target authentication data and management of subscription relations;
the data subsystem 21 is used for realizing the data storage and distribution functions; the stored data are as follows: (1) Data generated by an application running on the node and dependent data; (2) Shadow data of a moving target of the grid covered by the edge node; and (3) the own operation data of the node. Wherein the application relies on data, the data system needs to synchronously distribute the data related to the moving object to the adjacent edge nodes when the moving object will enter the coverage area of other edge nodes.
The second measurement subsystem 22 is used to discover moving objects (e.g., vehicles) that are driven into the covered area and establish a communication link therewith; also used for moving targets (e.g. velicle) to send their own velocity, acceleration, direction and position information to the edge node's measurement system;
the second application management subsystem 23 is used for managing the application life cycle according to the instruction of the control subsystem; the application on each edge node is responsible for providing services to a specific mobile object. There are two applications: one is to reside on the edge node for a long period of time, serving all moving targets within all grids hosted by the edge node; the other is a personalized application which is actively applied to the system by the mobile target, and the application serves for special service of the mobile target and is destroyed after being used up;
the control subsystem 24 is configured to predict the direction of travel of the moving object based on real-time motion data and map data measured by the measurement system for each moving object currently being serviced, estimate when (e.g., estimate the next location based on predicted velocity and newton's law of motion, and correct the predicted velocity based on feedback values for the actual measured velocity and location for the next location) to leave the grid range serviced by the current edge node, predict the probability of the object grid it is likely to go to, determine when to which of the object grids are associated with edge node synchronization data, and trigger the target edge node to create the desired application services for the moving object.
The schematic diagram of the system is shown in fig. 2, in which: s1-1 to S10-4 represent grids, S1-1 represents a network of first rows and first columns, S10-4 represents a grid of first rows and first columns, S1-1, S2-1, S3-1, S4-1, S1-2, S2-2, S3-2, S4-2, S1-3, S2-3, S3-3, S4-3, S1-4, S2-4, S3-4, S4-4 are areas covered by edge node 1, S4-1, S5-1, S6-1, S7-1, S4-2, S5-2, S6-2, S7-2, S4-3, S5-3, S6-3, S7-3, S4-4, S5-4, S6-4, S7-4 are areas covered by edge node 2, S7-1, S8-1, S9, S1-1, S6-3, S7-4, S7-3, S9-4, S10-3, S10-4 and S10-4. S4-1, S4-2, S4-3, S4-4 are the intersection areas covered by edge node 1 and edge node 2, and S7-1, S7-2, S7-3, S7-4 are the intersection areas covered by edge node 2 and edge node 3.
Wherein adjacent concepts are defined as follows:
the present system divides the area served into a number of grids, each edge node covering a number of grids, the grids covered by each edge node allowing for a portion of overlap. The covered grids are adjacent or crossed, i.e. the edge nodes can be considered as adjacent edge nodes. The edge nodes synchronize information of adjacent edge nodes from the central dispatching control system and automatically establish connection between the adjacent edge nodes.
According to another embodiment of the present invention, there is provided a management scheduling method for ubiquitous edge computing, including the following two cases:
in the first case, the target terminal T1 first enters the service area provided by the system of the present invention; as shown in fig. 3, when the target terminal T1 first enters the area covered by the present system, it is explained that no edge node provides service before, and at this time, T1 can only obtain data from the central scheduling control system, which specifically includes the following steps:
the central dispatching control system informs the current edge node of the target terminal T1 to provide service for the grid according to the grid where the target terminal T1 is positioned;
the target terminal T1 receives the notification information and automatically establishes connection with an edge node in the notification information;
the target terminal T1 reports measurement data to the connected edge node at regular time and uses the second measurement subsystem to measure;
the control subsystem in the edge node confirms that the mobile target enters for the first time according to the obtained measured value, and informs the data subsystem of synchronizing the ordering relation and the data on which the mobile target depends from the central dispatching control system;
after the data synchronization is completed, the control subsystem in the edge node inquires the application currently used by the target terminal T1 to the central dispatching control system, if the application does not exist, the second application management subsystem is informed to carry out application creation, and after the application is available, the control subsystem is used for informing the target terminal T1 of the application service address of the service T1;
in the area where the grids cross, when a plurality of edge nodes can provide service at the same time, T1 decides which edge area (namely the edge node covering the area where the current grid is located) is requested to be served according to the link quality between the edge nodes;
through the above process, the T1 terminal can continuously obtain service.
In the second case, as shown in fig. 4, the target terminal is driven into the edge node Mb coverage grid by the edge node Ma coverage grid, and the scheduling method includes the steps of:
the edge node Ma judges the probability of the target terminal entering other edge node coverage grids according to the running state of the current target terminal T1, and sends a notification message to the corresponding edge node when the probability exceeds pa. For example: the probability of driving in Mb is pa1, the probability of driving in Mc is pa2, pa1> pa sends notification to Mb informing Ma of the probability of driving in Mb by predicted T1, pa2< pa, and no notification is sent to Mc. The node of the last service T1 recorded by Mb is Ma, if M1 is used for serving T1 before Ma, the Ma needs to inform Mb, and M1 nodes are also used, the Mb records M1, and the Ma is used for serving T1, so that the purpose of recording is to realize the following data synchronous acceleration service;
when the second measurement subsystem Mb in the edge node finds that the target terminal enters the grid covered by the target terminal, namely, the measurement subsystem Ma sends a synchronous T1 signal, the target terminal is indicated to enter the edge node Mb area from the edge node Ma area;
when entering a grid crossing area, e.g. an Mb area by Ma, ma informs the terminal which edge nodes the area has to provide services in order for the terminal to establish a connection with these edge nodes;
the target terminal T1 reports measurement data to the connected edge node at regular time, reports the last edge node which provides service as an edge node Ma, and measures by utilizing a second measurement subsystem of the connected edge node;
the control subsystem in the edge node Mb judges the probability Pb1 that the target terminal T1 enters the edge node Mb area according to the driving direction and the speed of the target terminal, and when the predicted probability Pb11 (Pb 11=Pb1+w1+w2, wherein w1 and w2 are weight factors) exceeds a preset threshold Pb by combining the probability Pa1 in the edge node Ma notification message, the data subsystem is notified to synchronize the ordering relation of the target terminal T1 and the data relied by the mobile target from the edge node Ma or the central dispatching control system;
after the data synchronization is completed, the control subsystem in the edge node Mb inquires the control subsystem in the edge node Ma (when the control subsystem is driven in by the Ma) or the central scheduling control system about the application currently used by the target terminal T1, informs the second application management subsystem of creating the application, and informs the target terminal T1 of the completion of the application creation of the service target terminal T1 of the Mb region through the control subsystem after the application creation is completed;
in the area where the grids intersect, when there may be a plurality of edge nodes providing services at the same time, T1 decides which edge area to request services to, based on the link quality with the edge nodes.
Through the above process, the T1 terminal can continuously obtain service.
Furthermore, for moving targets, it is a difficulty when data is synchronized after edge node switching of the location occurs. If global data synchronization is performed, on one hand, the efficiency is low, and on the other hand, the economy is poor. The design of the invention is as follows:
(1) The design of gridding is adopted: the edge nodes cover a range of grids, and the grid ranges covered by a plurality of edge nodes are intersected. For example: the edge node 1 and the edge node 2 have a cross at s4, and when v1 enters s4, the edge node 1 decides to trigger the edge node 2 synchronous data according to the measurement data of v 1;
(2) A second measurement subsystem: the second measurement subsystem predicts the following action direction according to the measurement data of the moving target in time and the probability of entering a new grid.
The timely reported data of the moving target comprises the following steps: current velocity v1, acceleration a1, position data p;
measurement subsystem: calculating the speed v2 and the acceleration a2 calculated by the measuring system according to the position data p and the time;
(3) And a control subsystem: predicting the probability of the moving target entering the next grid according to the data provided by the measurement subsystem, wherein the prediction algorithm is as follows:
as shown in fig. 5, the grid on which the moving object is located is an a1b1 grid, the grid is a cross grid, and adjacent grids covered with adjacent other edge nodes are a1b2, a2b1, a2b2. In actual implementation, the probability estimation algorithm adopts a plug-in design, and different estimation algorithms are adopted according to the service requirements. The invention provides an example algorithm as follows:
when the time is t1, measuring the distance between the position and (a 1c, b2 c) according to the data reported by t1, wherein the distance between the position and (a 1c, b2 c) is d3t1, the distance between the position and (a 2c, b1 c) is d1t1, and the distance between the position and (a 2c, b2 c) is d2t1;
when the time is t2, measuring the distance between the position and (a 1c, b2 c) according to the data reported by t2, wherein the distance between the position and (a 2c, b1 c) is d3t2, the distance between the position and (a 2c, b1 c) is d1t2, and the distance between the position and (a 2c, b2 c) is d2t2;
the probability of entry is calculated as follows:
at time t 2: calculating the probability of entering the adjacent grid pd to be 1-d/sum (d) according to the direct distance, for example, the probability of entering a2b1 to be 1-d1t 1/(d1t1+d2t1+d3t1);
at time t 2: e1 =d1t2-d1t1, e2=d2t2-d2t1, e3=d3t2-d3t1;
based on the calculated difference from the previous time, the calculated probability pe (1-pm) -e/sum (e), where e excludes e <0, when e is negative, indicating that it is not possible to drive into the grid, indicating a small probability event, and when e <0, the sum of probabilities of the grid with probability pm,1-pm remaining as e >0 over a range, such as: e1<0, where e1 has a probability of pm, i.e., a probability of entering a2b1 is pm, a probability of entering a1b2 is (1-pm) -e3/sum (e2+e3), and pm has a minimum value of 0;
the probability of entering the corresponding grid is calculated as sum (wn×pet)/n+w0pd, wherein t takes n continuous periods, namely, the probability of the current distance is calculated after the corresponding weight pe of n continuous periods is taken, and the probability of the current entering the grid is obtained by the weight w 0. For example, when n is 3, the probability w3×pet3+w2×pet2+w1×pet1+w0×pd is obtained as the probability of the current driving in.
According to the measured value of the second measuring subsystem and the coverage condition of the grid, the control subsystem triggers the adjacent target edge node, informs the edge node to synchronize the dependent data of the moving target and the application ordering information of the moving target, and if the application needs to be created, informs the edge node to create a new application.
When a moving target enters the crossed edge node, the control node of the newly-entered edge region predicts the probability of actually entering the region according to the speed and the acceleration of the entering target, and false alarms of false entering and adjacent nodes are prevented.
(4) Data synchronization: the edge nodes all store edge node data, which is equivalent to that the edge nodes together form a distributed system, and the central control node stores Hash (Hash function) of each data block of index data. When each node synchronizes data, the adjacent node is prioritized to synchronize the Hash, and the data is desynchronized according to the Hash value. And if the adjacent node does not have data, synchronously indexing the adjacent node from the central node, searching the position of the data fragment corresponding to the Hash value, and downloading the data from the corresponding position.
As shown in fig. 6, 1) when T1 is driven into the area of edge node C by edge node b, edge node C requests data of which applications currently used by T1 from edge node b, if edge node b returns a fragment hash of T1 to edge node C and informs edge node C of other edge nodes that provide services for T1, such as edge node a;
2) The edge node c initiates application data synchronization used by T1 to the edge node a and the edge node c, and initiates a data synchronization request to the central dispatching control system when the same data synchronization is lost;
3) By the method, each edge node stores the data required by the object served by the edge node, global full-quantity storage is not needed, and the complexity of data storage can be reduced. When a moving object leaves the covered area for a period of time, for example, 3 days, the data of the moving object can be released, and the purpose of the immediate release is not to be because the moving object can have round trip, and when the object reappears, the overhead of data synchronization can be reduced.
In order to facilitate understanding of the above technical solutions of the present invention, the following describes specific embodiments of the present invention in a practical process.
(1) The user plays cloud games on the high-speed rail through the mobile phone, because tasks such as game rendering and the like are completed remotely, the mobile phone end only performs control use, and according to the existing scheme, the user cannot move at all and cannot experience on the high-speed rail. According to the embodiment, the edge nodes are deployed along the line, and the rendering service required by the cloud game can be continuously created through the system, so that smooth game experience service is provided for users.
(2) V2X (Vehicle to everything): when the cross-regional V2X service (namely the Internet of vehicles service) is adopted, service delay and even unavailability can be caused due to the cross-regional data intercommunication problem.
In summary, the key technical points of the invention are as follows:
(1) According to the gridding design of the edge node, when the moving target enters the cross grids, the probability estimation is carried out to trigger the adjacent grids to carry out data synchronization, so that services can be provided for the moving target entering the new grids more quickly, and smooth services can be provided for users
(2) The networking architecture provided by the invention records who the last edge node serves is when the mobile object passes through one edge node, so that when the mobile object needs to reach a new edge node, the system can know which edge nodes have previously served the object, and thus, when data is synchronized, the system can more efficiently know which edge nodes can obtain data, and can rapidly perform concurrent downloading
(3) The measurement services provided by the invention, and the design of probability estimation based thereon, provide support for when data migration and application migration are performed. The probability estimation provided by the invention is only one scheme, and in the actual implementation process, the probability estimation module can be designed according to plug-in, and various probability estimation methods can be provided according to different applications.
(4) The data storage structure of the system has no single node fault, and theoretically, each node which is recently served by each mobile target has more complete data than the last service node, and the service provided for the mobile target is not completely influenced by the fault of a certain node.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (10)
1. The management scheduling system for ubiquitous edge calculation is characterized by comprising a control layer and an edge layer, wherein the control layer comprises a central scheduling control system, and the edge layer automatically forms a mesh network by edge nodes to form an edge cloud;
the central dispatching control system comprises a dispatching subsystem, a first application management subsystem, a data distribution subsystem, a first measurement subsystem, an edge node management subsystem and a mobile target management subsystem;
the edge node comprises a data subsystem, a second measurement subsystem, a second application management subsystem and a control subsystem;
the scheduling subsystem is used for determining data management and application life cycle management of each edge node according to the monitoring information of each edge node;
the first application management subsystem is used for providing life cycle management of the global application and deciding to provide corresponding application services for the mobile target according to the ordering relation of the mobile target;
the data distribution subsystem is used for providing data slicing indexes, and each edge node decides a place for downloading data according to the slicing indexes;
the first measurement subsystem is used for measuring global speed measurement and helping a mobile target to select to access an edge service system; when a moving target initiates speed measurement, a speed measurement system of a control plane estimates a conforming target edge service system list according to the network time delay and the moving direction of the moving target, and a system which is recommended most preferentially is placed at the first place;
the edge node management subsystem is used for providing management and query services of each edge node, namely setting a grid range of the edge node service and an access mode of the edge node;
the mobile target management subsystem is used for managing the authentication data of the mobile target and the subscription relation;
the data subsystem is used for realizing the functions of data storage and distribution;
the second measurement subsystem is used for finding a moving target entering the covered area and establishing a communication link with the moving target; the method is also used for receiving the speed, acceleration, direction and position information of the moving target;
the second application management subsystem is used for managing the application life cycle according to the instruction of the control subsystem;
the control subsystem is used for predicting the advancing direction of the moving target according to the real-time motion data and the map data of each moving target which are currently served and obtained by the measurement of the second measurement subsystem, and estimating the probability of when the moving target leaves the grid range served by the current edge node and is expected to go to the target grid; but also to decide when to synchronize data to the edge nodes associated with which target grids and to trigger the target edge nodes to create the application services required by the moving target.
2. The system according to claim 1, wherein adjacent edge nodes in the edge layer are automatically connected to each other and form a mesh network.
3. The system of claim 1, wherein all data of the entire data management and scheduling system is stored in the edge node, and the central scheduling control system is also tied with global data.
4. The system of claim 1, wherein the data stored in the data subsystem includes data and dependent data generated by an application running on the edge node, shadow data of a moving object of a grid covered by the edge node, and running data of the edge node itself.
5. The ubiquitous edge computing managed scheduling system of claim 4, wherein the data subsystem is further configured to synchronously distribute moving object related data to neighboring edge nodes when the moving object enters other edge node coverage areas.
6. The system of claim 1, wherein the application is responsible for providing services to specific moving targets, wherein the application comprises two of:
the first is to reside on the edge node for a long time, serving all moving targets within all grids hosted by the edge node;
the second is a personalized application actively applied to the system by the mobile target, which is used for serving the special service of the mobile target.
7. A management scheduling method of ubiquitous edge computing, based on the implementation of the management scheduling system of ubiquitous edge computing as claimed in any one of claims 1-6, characterized in that the scheduling method is used for scheduling when a target terminal first enters a service area provided by the management scheduling system of ubiquitous edge computing, and the scheduling method comprises the following steps:
the central dispatching control system informs the target terminal of the edge node which provides service for the grid according to the grid where the target terminal is located;
the target terminal receives the notification information and automatically establishes connection with an edge node in the notification information;
the target terminal reports measurement data to the connected edge node at regular time, and the second measurement subsystem is utilized for measurement;
the control subsystem in the edge node confirms that the mobile target enters for the first time according to the obtained measured value, and informs the data subsystem of synchronizing the ordering relation and the data on which the mobile target depends from the central dispatching control system;
after the data synchronization is completed, the control subsystem in the edge node inquires the application currently used by the target terminal to the central dispatching control system, if the application does not exist, the second application management subsystem is informed to carry out application creation, and after the application is available, the control subsystem is used for informing the target terminal of the application service address of the service target terminal;
and the target terminal requests the service from the edge area through the corresponding application service address, so that the target terminal obtains continuous service.
8. The method for managing and scheduling ubiquitous edge computing according to claim 7, wherein said target terminal requests service from an edge area through an application service address corresponding thereto, further comprising the steps of:
in the grid crossing area, when a plurality of edge nodes provide service at the same time, the target terminal selects a corresponding edge area according to the link quality of the edge nodes and requests the service from the edge area.
9. A management scheduling method of ubiquitous edge computing, implemented based on the management scheduling system of ubiquitous edge computing according to any one of claims 1-6, characterized in that the scheduling method is used for scheduling when a target terminal is driven into an edge node Mb coverage grid by an edge node Ma coverage grid, the scheduling method comprising the steps of:
the edge node Ma judges the probability of the target terminal entering other edge node coverage grids according to the running state of the current target terminal, and sends a notification message to the corresponding edge node when the probability exceeds a preset value;
the second measurement subsystem of the edge node Mb finds that the target terminal enters the grid covered by the target terminal, and indicates that the target terminal enters the edge node Mb area from the edge node Ma area;
when entering a grid crossing area, the edge node at the previous moment informs the target terminal of the edge node which provides service for the area, and the target terminal is connected with the edge node which provides service for the area;
the target terminal reports measurement data to the connected edge node at regular time, reports the last edge node providing service as an edge node Ma, and measures by utilizing a second measurement subsystem of the connected edge node;
the control subsystem in the edge node Mb judges the probability of the target terminal entering the edge node Mb area according to the driving direction and the speed of the target terminal, and when the predicted probability exceeds a preset threshold value, the control subsystem informs the data subsystem of synchronizing the ordering relation of the target terminal and the data relied by the moving target from the edge node Ma or the central dispatching control system in combination with the probability in the edge node Ma notification message;
after the data synchronization is completed, the control subsystem in the edge node Mb inquires the control subsystem in the edge node Ma or the central scheduling control system about the application currently used by the target terminal, informs the second application management subsystem of creating the application, and informs the target terminal of the completion of the application creation of the service target terminal in the edge node Mb area through the control subsystem after the application creation is completed;
and the target terminal requests the service from the edge area through the corresponding application service address, so that the target terminal obtains continuous service.
10. The method for managing and scheduling ubiquitous edge computing according to claim 9, wherein said target terminal requests service from an edge area through an application service address corresponding thereto, further comprising the steps of:
in the grid crossing area, when a plurality of edge nodes provide service at the same time, the target terminal selects a corresponding edge area according to the link quality of the edge nodes and requests the service from the edge area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310037907.4A CN116132443B (en) | 2023-01-10 | 2023-01-10 | Management scheduling system and method for ubiquitous edge computing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310037907.4A CN116132443B (en) | 2023-01-10 | 2023-01-10 | Management scheduling system and method for ubiquitous edge computing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116132443A CN116132443A (en) | 2023-05-16 |
| CN116132443B true CN116132443B (en) | 2024-03-01 |
Family
ID=86302333
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310037907.4A Active CN116132443B (en) | 2023-01-10 | 2023-01-10 | Management scheduling system and method for ubiquitous edge computing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116132443B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633138A (en) * | 2019-08-28 | 2019-12-31 | 中山大学 | Automatic driving service unloading method based on edge calculation |
| CN112466158A (en) * | 2020-11-26 | 2021-03-09 | 东南大学 | Plane intersection-oriented vehicle collision risk assessment and prediction method |
| CN115022601A (en) * | 2022-07-25 | 2022-09-06 | 南京昊华科技有限公司 | IOT deep learning intelligent early warning research and judgment system and method based on big data |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11234204B2 (en) * | 2019-02-12 | 2022-01-25 | Intel Corporation | Server selection for vehicle communications and applications |
| US11473917B2 (en) * | 2020-07-14 | 2022-10-18 | Argo AI, LLC | System for augmenting autonomous vehicle perception using smart nodes |
-
2023
- 2023-01-10 CN CN202310037907.4A patent/CN116132443B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633138A (en) * | 2019-08-28 | 2019-12-31 | 中山大学 | Automatic driving service unloading method based on edge calculation |
| CN112466158A (en) * | 2020-11-26 | 2021-03-09 | 东南大学 | Plane intersection-oriented vehicle collision risk assessment and prediction method |
| CN115022601A (en) * | 2022-07-25 | 2022-09-06 | 南京昊华科技有限公司 | IOT deep learning intelligent early warning research and judgment system and method based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116132443A (en) | 2023-05-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8185159B2 (en) | Wireless communication system | |
| CN108777852A (en) | A kind of car networking content edge discharging method, mobile resources distribution system | |
| CN105554120A (en) | Data transmission method and data transmission device for vehicle communication, road side units | |
| KR20090017727A (en) | The method, apparatus and system for network resource reservation considering terminal speed and mobile terminal | |
| CN116261120B (en) | V2X service dynamic migration method and system based on cloud edge end cooperation | |
| JP2020535678A (en) | Base station switching method and equipment based on mobile device movement status | |
| US7395071B2 (en) | System and method of making location updating management on a mobile station, mobile station and mobile network | |
| CN116132443B (en) | Management scheduling system and method for ubiquitous edge computing | |
| CN113784286A (en) | Paging method, apparatus, storage medium and communication system | |
| CN114401270B (en) | Application migration method, system, device, electronic equipment and storage medium | |
| Derder et al. | An online target tracking protocol for vehicular ad hoc networks | |
| CN114124736A (en) | Network function virtualization in ad hoc groups | |
| CN108738027B (en) | Network processing method, resource management system and network equipment | |
| EP2963984B1 (en) | A method for resource allocation in a telecommunication system | |
| CN108174395B (en) | Base station cache management method and system based on transfer action evaluation learning framework | |
| CN115333604B (en) | Satellite terminal monitoring method, system, device and storage medium | |
| Mukherjee et al. | DAS: An intelligent three dimensional cost effective movement prediction of active users in heterogeneous mobile network | |
| WO2023052010A1 (en) | Trajectory data collection in mobile telecommunication systems | |
| Ali et al. | A survey of services placement mechanisms for future mobile communication networks | |
| CN114630295A (en) | Method and apparatus for autonomous fleet processing using broadcast guidance | |
| CN116419323A (en) | Service migration control method, device and system | |
| CN114449640A (en) | Data synchronization method, information transmission method and equipment | |
| CN104219641A (en) | Information transmitting method and device | |
| JP2022502898A (en) | How to reallocate data processing peripheral servers | |
| Han et al. | Population-based call admission control in wireless cellular networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |