CN114090221A - Dual migration method for tasks in edge computing environment - Google Patents
Dual migration method for tasks in edge computing environment Download PDFInfo
- Publication number
- CN114090221A CN114090221A CN202210068365.2A CN202210068365A CN114090221A CN 114090221 A CN114090221 A CN 114090221A CN 202210068365 A CN202210068365 A CN 202210068365A CN 114090221 A CN114090221 A CN 114090221A
- Authority
- CN
- China
- Prior art keywords
- user equipment
- edge server
- base station
- server
- information
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4812—Task transfer initiation or dispatching by interrupt, e.g. masked
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5077—Logical partitioning of resources; Management or configuration of virtualized resources
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45575—Starting, stopping, suspending or resuming virtual machine instances
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/502—Proximity
Landscapes
- Engineering & Computer Science (AREA)
- 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
A dual migration method of tasks in an edge computing environment belongs to the technical field of resource allocation and comprises the following steps: step S1, establishing an edge computing platform; step S2, the user device transmits heartbeat information to the edge server connected with the user device in communication; step S3, the edge server actively detects whether the ue logs out; when the edge server detects that the user equipment logs out, executing interrupt state migration; otherwise, carrying out the next step; step S4, the edge server actively detects whether the ue moves: when the edge server detects that the user equipment moves, executing continuous state transition; otherwise, the process returns to step S3. According to the scheme, the double migration method is used, the interrupt state migration and the continuous state migration of the user are considered, the user equipment is in communication connection with the nearest server, the transmission path between the server and the user is shortened, the task is preloaded in the nearest server, and therefore the cloud service quality of the task is guaranteed.
Description
Technical Field
The invention belongs to the technical field of resource allocation, and particularly relates to a dual migration method of tasks in an edge computing environment.
Background
With the development of cloud technology, users can log in to the network and enjoy cloud services at any time in different places. The edge computing reduces access delay, saves bandwidth and improves throughput by deploying tasks at positions close to access points such as base stations. However, earlier edge computing architectures only distributed the same server to face all users, regardless of the transformation of the users' actual geographic locations. When the user is in a mobile state, the distance between the user and the server becomes longer and longer, which causes the network transmission path to become longer and the network delay to increase, so that the service quality of the whole cloud computing is reduced.
In addition, chinese patent application publication No. CN111274031A discloses a method and an apparatus for dynamic migration authentication of edge service in end-cloud cooperation. When the edge service data migration authentication under terminal movement needs to be achieved, a first edge sends a notification for service migration and timestamp information to a mobile terminal, then the mobile terminal obtains a second edge capable of providing services through a cloud end, the mobile terminal establishes connection with the second edge, the mobile terminal notifies the first edge to execute migration of the services to the second edge, the first edge executes service migration to the second edge, the second edge sends migration completion confirmation information to the mobile terminal, and the second edge and the mobile terminal establish edge services. Under the scene of edge computing service, the mobile terminal can safely and quickly realize the authentication of service data migration between different edge servers. There are two disadvantages in this solution:
according to the technical scheme, in the service forward moving process, the mobile terminal and the edge terminal are always in a communication connection state. In a real use environment, the mobile terminal is offline after communicating with the first edge, and after a period of time, the mobile terminal is online in other areas and is in communication connection with the second edge, and at this time, the second edge provides continuous cloud service for the user. The above patent scheme cannot satisfy cloud task migration after communication interruption in a real environment.
Disclosure of Invention
In view of the foregoing deficiencies of the prior art, it is an object of the present invention to provide a method for dual migration of tasks in an edge computing environment.
A method of dual migration of tasks in an edge computing environment, comprising the steps of:
step S1, establishing an edge computing platform: the edge computing platform comprises at least 2 area networks which are in communication connection with a cloud end; each regional network is provided with an optical fiber backbone; the optical fiber backbone is connected with 1 storage area network and at least 2 edge servers; each edge server in the local area network has access to storage on the storage area network; at least 2 storage servers are arranged in the storage area network; the edge server is deployed near a base station;
step S2, the user equipment periodically transmits heartbeat information to the edge server in communication connection with the user equipment, where the heartbeat information includes: the identification number, latitude, longitude, base station name and local area network address of the current connection line and heartbeat sending time of the user equipment;
step S3, the edge server actively detects whether the ue logs out; when the edge server detects that the user equipment logs out, executing interrupt state migration, putting task information of the user equipment in storage area network equipment corresponding to the edge server which the user equipment possibly logs in by predicting the frequently-occurring geographic position of the user equipment, and enabling the user equipment to access cloud services from the nearest edge server after the user equipment logs in again; otherwise, carrying out the next step;
step S4, the edge server actively detects whether the ue moves: when the edge server detects that the user equipment moves, executing continuous state migration, continuously using cloud service while the user moves, always maintaining the login state, performing task information conversion and connection channel conversion through a handoff program, migrating the task information of the user equipment to a storage server corresponding to the edge server nearest to the user, and providing the cloud service for the user equipment by the edge server nearest to the user equipment; otherwise, the process returns to step S3.
Further, step S3 further includes:
step S301, task information conversion: the cloud acquires all heartbeat information of the user equipment from the edge server, calculates login probability of the user equipment in each edge server coverage area, and dispersedly stores task information of the user equipment in the storage servers corresponding to the edge servers according to the login probability, wherein at the moment, the probability that the user equipment logs in a platform in the edge server coverage area is equal to the percentage of task information quantity of the storage servers corresponding to the edge servers in the total quantity of the task information;
step S302, connection channel conversion: the user equipment sends a login request of the cloud system; the base station which establishes communication connection with the user equipment guides the login request to the edge server group, and finds the nearest edge server as a target server by using an arbitrary broadcasting mode;
when the target server receives a login request of any broadcasting base station, the target server executes a virtual machine starting program and loads task information of user equipment, responds to the login request of the user equipment, and sends a response packet containing a unicast address of the target server to the user equipment; after the user equipment establishes a connection channel with the target server, information is transmitted between the user equipment and the target server in a unicast mode.
Further, step S4 further includes:
step S401: the edge server actively detects whether the user equipment is displaced; the edge server receives heartbeat information periodically transmitted by the user equipment, and judges whether the user equipment moves according to the local area network address position of the user equipment connecting base station: when the user equipment moves, the connected base station is switched, and the local area network address of the connected base station is changed, which indicates that the user equipment has moved; when the user equipment is displaced, determining a new base station according to the local area network address of the connected base station;
step S402: the old base station initiates a hand-off program to the new base station, the old base station sends switching request information to the new base station, and the switching request information comprises a new user equipment identification number; then, the new base station replies switching confirmation information to inform the old base station of the identification number and the access channel of the new user equipment;
at the moment, the old base station still provides the cloud service of the user equipment; step S403 and step S406 are performed simultaneously;
step S403: the old base station sends migration information to the old edge server, wherein the migration request information comprises information of the new base station, the service type used by the user equipment and the required virtual hardware resource; a new base station establishing communication connection with the user equipment finds a nearest edge server as a new edge server in an arbitrary broadcasting mode;
step S404: the old edge server sends migration request information to the new edge server, and the migration request information informs the new edge server of the service type used by the user equipment and the required virtual hardware resources; the new edge server returns migration confirmation information to the old edge server according to the migration request information after completing the new virtual machine starting program;
step S405: at the moment, the user equipment still obtains the cloud service from the old edge server through the old base station, so that the old edge server calls the task information currently used by the user equipment in the corresponding storage server, simultaneously forwards transmits the task information to the new edge server, and keeps the task information in the storage server corresponding to the new edge server to finish task information conversion; then go to step S407;
step S406: after receiving the switching confirmation information, the old base station forwards transmits wireless reconstruction information to the user equipment, wherein the wireless reconstruction information comprises a new user equipment identification number and an access channel; the old base station transmits the identification number transfer information to the new base station; then go to step S407;
step S407: starting a handoff procedure of a second layer, wherein the old base station and the new edge server simultaneously forward transmit the temporary storage data to the new base station, and the new edge server and the old edge server are synchronized;
step S408: the user equipment sends wireless connection completion information to the new base station;
step S409: the new base station transmits the temporary storage data to the user equipment, a connection channel is established between the new base station and the user equipment, the connection channel conversion is completed, and the new edge server provides the cloud service for the user equipment.
The scheme adopts a double migration method, and considers the interrupt state migration and the continuous state migration of the user, and the idea is that no matter where and when the user logs in the cloud service, the cloud service is accessed by the server nearest to the user. According to the scheme, on one hand, through continuous monitoring of the geographic position of the user, the user equipment can be in communication connection with the nearest server no matter where and when the user uses the cloud service, the transmission path between the server and the user is shortened, and the transmission delay time is effectively reduced.
And (3) interrupted state migration: and when the user moves to a different place and logs in the cloud system again, the user continues to use the previous cloud service. In the traditional scheme, because the information of the previous cloud service is stored in the previous server, the cloud distributes the previous server to the user for use no matter where the user logs in for the second time, and when the previous server is far away from the user, the high-quality cloud service is difficult to maintain. According to the scheme, the user equipment logs out of the cloud system, and task information of the user equipment is placed in storage area network equipment corresponding to an edge server which the user equipment possibly logs in by predicting the frequently-occurring geographic position of the user equipment, so that the time for loading the private data of the user by the server next time is shortened. And find out the nearest server of user through broadcasting or mediating the server arbitrarily, let the customer equipment access the cloud end service from the nearest marginal server, shorten the transmission path between user and the server.
And (3) continuous state transition: the user continuously uses the cloud service while moving, the cloud service is always kept in a login state, the current position of the user is mainly concerned, task information conversion and connection channel conversion are carried out through a hand-off program, the task information of the user equipment is migrated to a storage server corresponding to an edge server nearest to the user, and the edge server nearest to the user provides the cloud service for the user equipment. The synchronization of the task information is added in the hand-off program, so that the new edge server and the old edge server are synchronized, the service interruption time is shortened, and the new edge server and the old edge server can be switched in a short time. The user equipment is always in butt joint with the edge server which is closest to the user equipment, and the distance between the user equipment and the edge server is always one hop number.
Drawings
FIG. 1 is a schematic structural view of the present invention;
FIG. 2 is a flow chart of the continuous state transition of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
A method of dual migration of tasks in an edge computing environment, comprising the steps of:
step S1: establishing an edge computing platform: as shown in fig. 1, the edge computing platform includes at least 2 area networks each communicatively connected to a cloud; each regional network is provided with an optical fiber backbone; the optical fiber backbone is connected with 1 storage area network and at least 2 edge servers; at least 2 storage servers are arranged in the storage area network; the edge server is deployed near a base station.
The edge server of the scheme uses Virtual Machine (Virtual Machine) technology to divide one server into a plurality of Virtual machines, so as to share hardware, software licenses and management resources with maximized efficiency.
Storage Area Network (SAN): and the edge server is connected with the storage space pool to provide a shared storage space pool for the edge server. Each edge server in the local area network can access the storage on the storage area network, and the storage servers in the storage area network are independent and communicated with each other, so that the equipment expansion and the data backup are facilitated.
A storage server: is a server connected to the storage area network and having data storage function.
Step S2: the user equipment periodically transmits heartbeat information to an edge server in communication connection with the user equipment, wherein the heartbeat information comprises: the identification number, latitude, longitude, base station name and local area network address of the current connection line and heartbeat sending time of the user equipment.
According to the scheme, the information of the user equipment is transmitted to the edge server, and the edge server grasps the movement of the user equipment and adjusts service deployment and the like. Compared with a centralized management mode of transmitting the information of the user equipment to the cloud, the scheme reduces network delay. Compared with an autonomous management mode that the information of the user equipment is transmitted to the edge server or the cloud end when needed, the scheme can systematically master the movement of the user equipment and adjust the requirements of the user in real time.
The scheme adopts a heartbeat transmission mechanism, so that the edge server can master the information and the movement of the user equipment. The edge server receives the heartbeat information, records and analyzes the heartbeat information, and if the user equipment logs out, the interrupt state migration is carried out; and if the user equipment is displaced, carrying out continuous state transition.
Step S3: the edge server actively detects whether the user equipment logs out; and when the edge server detects that the user equipment logs out, executing interrupt state migration, and otherwise, executing the next step.
And the interrupt state is migrated, and the task information of the user equipment is migrated to the storage server closest to the user equipment, so that the time for loading the task information from the storage area network by the edge server is reduced, and the closest edge server provides cloud service for the user equipment to shorten a transmission path.
Interrupting state migration, and the process is as follows:
step S301: and (3) task information conversion: the cloud end acquires all heartbeat information of the user equipment from the edge server, calculates login probability of the user equipment in coverage areas of all the edge servers, and dispersedly stores task information of the user equipment in the storage servers corresponding to the edge servers according to the login probability.
According to the task information conversion method, after the user equipment logs out, the next login position of the user equipment is predicted, and the task information is distributed to the storage servers in the storage area network according to the login probability, so that the task information quantity of each storage server is different, and the storage server corresponding to the edge server in the area with the highest login probability stores the task information with the highest proportion, so that the transmission path of the task information is shortened, the data quantity downloaded by the edge server from other devices far away in the storage area network is reduced, and the service quality of edge calculation is guaranteed.
And analyzing and comparing the scheme with the mode of evenly distributing the task information.
In the task information average allocation method, the average task information migration amount D when the user equipment logs in from any edge server:
and M represents the total amount of task information of the user equipment.
N, representing the total number of edge servers.
Pi represents the probability that the user equipment logs in the platform in the coverage area of the ith edge server, and i is more than or equal to 1 and less than or equal to N.
After the task information is converted, the average task information migration amount Dm when the user equipment logs in from any edge server is as follows:
di represents the percentage of the amount of task information of the storage server corresponding to the ith edge server to the total amount of task information.
According to the scheme, the task information of the user equipment is dispersedly stored in the storage server corresponding to the edge server according to the login probability Pi, namely Pi = Di. Therefore, the temperature of the molten metal is controlled,
the corresponding mode of the edge server and the storage server corresponding to the edge server adopts a distance mode, namely, the storage server closest to the edge server is the storage server corresponding to the edge server by default.
When the user equipment logs in the ith edge server, the difference value Xi between the amount of the task information required to be downloaded in the task information average distribution mode and the amount of the task information required to be downloaded in the task information conversion of the scheme is as follows:
when in use
And when the number Xi is positive, the amount of the task information required to be downloaded is large, which indicates that the task information is distributed evenly, and at the moment, the task information conversion of the scheme improves the transmission efficiency of the task information. On the contrary, when
In the case of the present embodiment, Xi is a negative number, which means that the amount of the downloaded task information required for the task information average distribution method is small, but Pi is small at this time, so that the loss caused by the task information conversion in the present embodiment is not large. Therefore, overall, the task information of the present solution is better than the task information average distribution method in terms of the transmission efficiency of the task information.
For example, the task information can be transmitted to the server. In this example, there are 3 edge servers, edge server a, edge server b, and edge server c; the user equipment logs in for 4 times in a mobile mode, and the logged edge servers are respectively as follows: edge server a, edge server b, edge server a.
In the method of evenly distributing task information, since the amount of task information per edge server is 1/3, the required task information migration amount D is 2/3 for each registration regardless of which edge server the user registers.
According to the scheme, the task information is converted, and when the user equipment logs in for the first time, all task information quantity is stored in the storage server corresponding to the edge server a. When the user equipment logs in for the second time, the geographic position is different from that of the user equipment logged in for the first time, and at this time, the probability of the user equipment appearing in the two geographic positions is 1/2, so when the user equipment logs out, the task information amount is divided into two parts on average and stored in the storage server corresponding to the edge server a and the storage server corresponding to the edge server b respectively, and at this time, when the user equipment logs in any edge server for the third time, the average task information migration amount Dm is 4/9. Assuming that the geographical location of the user equipment logged in for the third time corresponds to the edge server a, it is found that the storage server corresponding to the edge server a already stores 1/2 task information amount, and only needs to download 1/2 task information amount. When the user device logs out again, edge server a is 2/3, edge server b is 1/3, and edge server c is 0 for the login probability; and distributing the task information according to the login probability, wherein the average task information migration quantity Dm is 3/8 when the user equipment logs in any edge server for the fourth time.
Therefore, Dm is smaller than D, and the task information migration amount required by the task information conversion of the scheme is small.
| Check-in Number of times | Edge server a, ren Traffic information volume | Edge server b, ren Traffic information volume | Edge server c, ren Traffic information volume | Means for evenly distributing task information, averaging Task information migration volume D | Task information conversion, mean task Traffic information migration amount Dm |
| 1 | 1☆ | 0 | 0 | 2/3 | 0 |
| 2 | 1/2 | 1/2☆ | 0 | 2/3 | 1/2 |
| 3 | 2/3☆ | 1/3 | 0 | 2/3 | 4/9 |
| 4 | 3/4☆ | 1/4 | 0 | 2/3 | 3/8 |
Table 1 is an example of task information transformation for this scheme (it indicates the edge server that logs in).
In this embodiment, the moving range of the ue may be limited to a local area network (e.g., an internal arrow in the left-side local area network in fig. 1) to implement the task information flow inside the local area network, or may also implement the task information transmission in a wide area network across the local area network (e.g., an arrow between the left-side local area network and the right-side local area network in fig. 1).
Step S302: connection channel conversion: the user equipment sends a login request of the cloud system; and the base station establishing communication connection with the user equipment guides the login request to the edge server group, and finds the nearest edge server as a target server in an arbitrary broadcasting mode.
When the target server receives a login request of any broadcasting base station, the target server executes a virtual machine starting program and loads task information of user equipment, responds to the login request of the user equipment, and sends a response packet containing a unicast address of the target server to the user equipment; after the user equipment establishes a connection channel with the target server, information is transmitted between the user equipment and the target server in a unicast mode.
And (4) randomly broadcasting: any sender in a group corresponds to a communication between several nearest receivers in the topology. Whenever a packet is sent to the anycast address, it is sent to the nearest base station providing internet access. The scheme adopts a universal random broadcasting mode and is not shown.
Step S4: the edge server actively detects whether the user equipment moves: when the edge server detects the ue movement, it performs continuous state transition, otherwise, it returns to step S3.
And (3) continuous state transition: the user continuously uses the cloud service while moving, the cloud service is always kept in a login state, the current position of the user is mainly concerned, task information conversion and connection channel conversion are carried out through a hand-off program, the task information of the user equipment is migrated to a storage server corresponding to an edge server nearest to the user, and the edge server nearest to the user equipment provides the cloud service for the user equipment.
Referring to fig. 2, step 4 is further analyzed.
Step S401: the edge server actively detects whether the user equipment is displaced; the edge server receives heartbeat information periodically transmitted by the user equipment, and judges whether the user equipment moves according to the local area network address position of the user equipment connecting base station: when the user equipment moves, the connected base station is switched, and the local area network address of the connected base station is changed, which indicates that the user equipment has moved; and when the user equipment is displaced, determining a new base station according to the local area network address of the connected base station.
Step S402: the old base station initiates a hand-off program to the new base station, the old base station sends switching Request information (Handover Request) to the new base station, and the switching Request information comprises a new user equipment identification number; then, the new base station replies a Handover acknowledgement message (Handover acknowledgement) to inform the old base station of the new ue identification number and access channel.
At this time, the old base station still provides the cloud service of the user equipment.
Step S403 and step S406 are performed simultaneously. The old base station will transfer the identification number and access channel of the new ue to the ue in step S406, thereby shortening the time of the handover procedure.
Step S403: the old base station sends Migration Information (Migration Information) to the old edge server, wherein the Migration request Information comprises Information of the new base station, the service type used by the user equipment and the required virtual hardware resource; and the new base station establishing communication connection with the user equipment uses an arbitrary broadcasting mode to find the nearest edge server as a new edge server.
Step S404: the old edge server sends Migration Request information (Migration Request) to the new edge server, and the Migration Request information informs the new edge server of the service type used by the user equipment and the required virtual hardware resources; and the new edge server returns Migration confirmation information (Migration acknowledgement) to the old edge server according to the Migration request information after finishing the new virtual machine starting program.
Step S405: at the moment, the user equipment still obtains the cloud service from the old edge server through the old base station, so that the old edge server calls the task information currently used by the user equipment in the corresponding storage server, simultaneously forwards transmits the task information to the new edge server, and keeps the task information in the storage server corresponding to the new edge server to finish task information conversion; then go to step S407.
Step S406: after receiving Handover acknowledgement information (Handover acknowledgement), the old base station forwards Radio Resource Control Connection Reconfiguration information to the user equipment, wherein the Radio Reconfiguration information comprises a new user equipment identification number and an access channel; the old base station transmits identification Number Transfer information (Sequence Number Status Transfer) to the new base station, and informs the new base station of identification Number state information of PDCP (packet Data conversion protocol) uploaded and downloaded by the user equipment; then go to step S407.
Step S407: and starting a handoff procedure of a second layer, wherein the old base station and the new edge server simultaneously forward the temporary storage data to the new base station, and the new edge server and the old edge server are synchronized to shorten the interruption time of the cloud service.
Step S408: the user equipment transmits Radio Resource Control Connection Reconfiguration Complete to the new base station.
Step S409: the new base station transmits the temporary storage data to the user equipment, a connection channel is established between the new base station and the user equipment, the connection channel conversion is completed, and the new edge server provides the cloud service for the user equipment.
In the continuous state transition process, when the second-layer handoff program is not entered, the user equipment still obtains the cloud service from the old edge server. The old edge server needs to forward the task information to the new edge server, and the new edge server does not provide the cloud service of the user equipment until the new edge server and the old edge server are synchronized.
According to the scheme, the synchronization of the task information is increased in the handoff program, so that the new edge server and the old edge server are synchronized, the service interruption time is shortened, and the new edge server and the old edge server can be switched in a short time. When the task information needing synchronization is small, the task information can be synchronized before the handoff procedure is finished, and the completion time of the handoff procedure is never prolonged. Theoretically, the handoff procedure is completed in 55 milliseconds, which is extremely short.
The NS2 is used for simulation, the optical fiber backbone bandwidth is set to be 1G, the radius of an access base station is 250m, the background flow rate is 100Mbps, and the influence of the scheme on the transmission delay time is tested.
In the conventional architecture, the ue still accesses from the original edge server regardless of the base station accessed by the ue, and thus, the number of hops between the ue and the edge server increases as the ue moves.
In the scheme, the user equipment is always in butt joint with the edge server closest to the user equipment, and the user equipment and the edge server are always separated by one hop number.
For the interrupt time: no matter the size of the task information which needs to be migrated by the user equipment, the old edge server provides the cloud service of the user equipment before the new edge server and the old edge server are synchronized, and the new edge server and the old edge server are switched to the new edge server after the new edge server and the old edge server are synchronized. The required interruption time does not change regardless of the size of the archive of task information that the user equipment needs to migrate. The interruption time is about 1RTT when the old server informs the user equipment through a message.
For the synchronization time: instead of migrating the entire virtual machine, we only need to synchronize the task information of the user devices, thus shortening the content of synchronization. When the task information is 1K, the synchronization time is 12.7 milliseconds; when the task information is 100M, the synchronization time is 26 seconds. It can be seen that as the task information increases, the synchronization time also increases. Of course, the bandwidth of the optical fiber backbone is increased, and the synchronization time can be effectively shortened.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (3)
1. A method for dual migration of tasks in an edge computing environment, comprising the steps of:
step S1, establishing an edge computing platform: the edge computing platform comprises at least 2 area networks which are in communication connection with a cloud end; each regional network is provided with an optical fiber backbone; the optical fiber backbone is connected with 1 storage area network and at least 2 edge servers; each edge server in the local area network has access to storage on the storage area network; at least 2 storage servers are arranged in the storage area network; the edge server is deployed near a base station;
step S2, the user equipment periodically transmits heartbeat information to the edge server in communication connection with the user equipment, where the heartbeat information includes: the identification number, latitude, longitude, base station name and local area network address of the current connection line and heartbeat sending time of the user equipment;
step S3, the edge server actively detects whether the ue logs out; when the edge server detects that the user equipment logs out, executing interrupt state migration, putting task information of the user equipment in storage area network equipment corresponding to the edge server which the user equipment possibly logs in by predicting the frequently-occurring geographic position of the user equipment, and enabling the user equipment to access cloud services from the nearest edge server after the user equipment logs in again; otherwise, carrying out the next step;
step S4, the edge server actively detects whether the ue moves: when the edge server detects that the user equipment moves, executing continuous state migration, continuously using cloud service while the user moves, always maintaining the login state, performing task information conversion and connection channel conversion through a handoff program, migrating the task information of the user equipment to a storage server corresponding to the edge server nearest to the user, and providing the cloud service for the user equipment by the edge server nearest to the user equipment; otherwise, the process returns to step S3.
2. The method for dual migration of tasks in an edge computing environment according to claim 1, wherein step S3 further comprises:
step S301, task information conversion: the cloud acquires all heartbeat information of the user equipment from the edge server, calculates login probability of the user equipment in each edge server coverage area, and dispersedly stores task information of the user equipment in the storage servers corresponding to the edge servers according to the login probability, wherein at the moment, the probability that the user equipment logs in a platform in the edge server coverage area is equal to the percentage of task information quantity of the storage servers corresponding to the edge servers in the total quantity of the task information;
step S302, connection channel conversion: the user equipment sends a login request of the cloud system; the base station which establishes communication connection with the user equipment guides the login request to the edge server group, and finds the nearest edge server as a target server by using an arbitrary broadcasting mode;
when the target server receives a login request of any broadcasting base station, the target server executes a virtual machine starting program and loads task information of user equipment, responds to the login request of the user equipment, and sends a response packet containing a unicast address of the target server to the user equipment; after the user equipment establishes a connection channel with the target server, information is transmitted between the user equipment and the target server in a unicast mode.
3. The method for dual migration of tasks in an edge computing environment according to claim 1 or 2, wherein step S4 further comprises:
step S401: the edge server actively detects whether the user equipment is displaced; the edge server receives heartbeat information periodically transmitted by the user equipment, and judges whether the user equipment moves according to the local area network address position of the user equipment connecting base station: when the user equipment moves, the connected base station is switched, and the local area network address of the connected base station is changed, which indicates that the user equipment has moved; when the user equipment is displaced, determining a new base station according to the local area network address of the connected base station;
step S402: the old base station initiates a hand-off program to the new base station, the old base station sends switching request information to the new base station, and the switching request information comprises a new user equipment identification number; then, the new base station replies switching confirmation information to inform the old base station of the identification number and the access channel of the new user equipment;
at the moment, the old base station still provides the cloud service of the user equipment; step S403 and step S406 are performed simultaneously;
step S403: the old base station sends migration information to the old edge server, wherein the migration request information comprises information of the new base station, the service type used by the user equipment and the required virtual hardware resource; a new base station establishing communication connection with the user equipment finds a nearest edge server as a new edge server in an arbitrary broadcasting mode;
step S404: the old edge server sends migration request information to the new edge server, and the migration request information informs the new edge server of the service type used by the user equipment and the required virtual hardware resources; the new edge server returns migration confirmation information to the old edge server according to the migration request information after completing the new virtual machine starting program;
step S405: at the moment, the user equipment still obtains the cloud service from the old edge server through the old base station, so that the old edge server calls the task information currently used by the user equipment in the corresponding storage server, simultaneously forwards transmits the task information to the new edge server, and keeps the task information in the storage server corresponding to the new edge server to finish task information conversion; then go to step S407;
step S406: after receiving the switching confirmation information, the old base station forwards transmits wireless reconstruction information to the user equipment, wherein the wireless reconstruction information comprises a new user equipment identification number and an access channel; the old base station transmits the identification number transfer information to the new base station; then go to step S407;
step S407: starting a handoff procedure of a second layer, wherein the old base station and the new edge server simultaneously forward transmit the temporary storage data to the new base station, and the new edge server and the old edge server are synchronized;
step S408: the user equipment sends wireless connection completion information to the new base station;
step S409: the new base station transmits the temporary storage data to the user equipment, a connection channel is established between the new base station and the user equipment, the connection channel conversion is completed, and the new edge server provides the cloud service for the user equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210068365.2A CN114090221B (en) | 2022-01-20 | 2022-01-20 | Dual migration method for tasks in edge computing environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210068365.2A CN114090221B (en) | 2022-01-20 | 2022-01-20 | Dual migration method for tasks in edge computing environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114090221A true CN114090221A (en) | 2022-02-25 |
| CN114090221B CN114090221B (en) | 2022-04-29 |
Family
ID=80308979
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210068365.2A Active CN114090221B (en) | 2022-01-20 | 2022-01-20 | Dual migration method for tasks in edge computing environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114090221B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5818448A (en) * | 1996-07-02 | 1998-10-06 | Sun Microsystems, Inc. | Apparatus and method for identifying server computer aggregation topologies |
| CN101841876A (en) * | 2009-03-20 | 2010-09-22 | 上海贝尔股份有限公司 | Transfer method and device between unauthorized mobile access network and honeycomb communication network |
| CN105812443A (en) * | 2014-12-31 | 2016-07-27 | 乐视网信息技术（北京）股份有限公司 | P2P system and P2P system business access method |
| CN109076357A (en) * | 2016-05-28 | 2018-12-21 | 华为技术有限公司 | Application method, relevant device and system are migrated in mobile limbic system |
| WO2020024978A1 (en) * | 2018-07-31 | 2020-02-06 | 中兴通讯股份有限公司 | Device, method, apparatus, and readable storage medium for virtual machine migration |
| CN111274031A (en) * | 2020-01-16 | 2020-06-12 | 国家电网有限公司信息通信分公司 | Method and device for dynamic migration authentication of edge service with cooperation of end and cloud |
| CN111491332A (en) * | 2020-04-20 | 2020-08-04 | 中国石油大学(北京) | Dynamic service migration method and device |
| CN112463337A (en) * | 2020-12-08 | 2021-03-09 | 内蒙古大学 | Workflow task migration method used in mobile edge computing environment |
| CN113163409A (en) * | 2021-03-16 | 2021-07-23 | 重庆邮电大学 | Mobile edge computing service placement strategy based on artificial intelligence |
| CN113301169A (en) * | 2021-07-23 | 2021-08-24 | 杭州雅观科技有限公司 | Edge network switching method based on dynamic mobile device behavior prediction |
| CN113364626A (en) * | 2021-06-08 | 2021-09-07 | 国网江苏省电力有限公司 | Service placement and bandwidth allocation method for video analysis application facing edge environment |
| CN113852693A (en) * | 2021-09-26 | 2021-12-28 | 北京邮电大学 | Migration method of edge computing service |
-
2022
- 2022-01-20 CN CN202210068365.2A patent/CN114090221B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5818448A (en) * | 1996-07-02 | 1998-10-06 | Sun Microsystems, Inc. | Apparatus and method for identifying server computer aggregation topologies |
| CN101841876A (en) * | 2009-03-20 | 2010-09-22 | 上海贝尔股份有限公司 | Transfer method and device between unauthorized mobile access network and honeycomb communication network |
| CN105812443A (en) * | 2014-12-31 | 2016-07-27 | 乐视网信息技术（北京）股份有限公司 | P2P system and P2P system business access method |
| CN109076357A (en) * | 2016-05-28 | 2018-12-21 | 华为技术有限公司 | Application method, relevant device and system are migrated in mobile limbic system |
| WO2020024978A1 (en) * | 2018-07-31 | 2020-02-06 | 中兴通讯股份有限公司 | Device, method, apparatus, and readable storage medium for virtual machine migration |
| CN111274031A (en) * | 2020-01-16 | 2020-06-12 | 国家电网有限公司信息通信分公司 | Method and device for dynamic migration authentication of edge service with cooperation of end and cloud |
| CN111491332A (en) * | 2020-04-20 | 2020-08-04 | 中国石油大学(北京) | Dynamic service migration method and device |
| CN112463337A (en) * | 2020-12-08 | 2021-03-09 | 内蒙古大学 | Workflow task migration method used in mobile edge computing environment |
| CN113163409A (en) * | 2021-03-16 | 2021-07-23 | 重庆邮电大学 | Mobile edge computing service placement strategy based on artificial intelligence |
| CN113364626A (en) * | 2021-06-08 | 2021-09-07 | 国网江苏省电力有限公司 | Service placement and bandwidth allocation method for video analysis application facing edge environment |
| CN113301169A (en) * | 2021-07-23 | 2021-08-24 | 杭州雅观科技有限公司 | Edge network switching method based on dynamic mobile device behavior prediction |
| CN113852693A (en) * | 2021-09-26 | 2021-12-28 | 北京邮电大学 | Migration method of edge computing service |
Non-Patent Citations (2)
| Title |
|---|
| QINGLAN PENG，YUNNI XIA，ZENG FENG，JIA LEE，CHUNRONG WU,XIN LUO: "Mobility Aware and Migration Enabled Online Edge User Allocation in Mobile Edge Computing", 《2019 IEEE INTERNATIONAL CONFERENCE ON WEB SERVICES》 * |
| 蒋椿磊: "移动边缘计算中基于移动代理的任务迁移研究", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114090221B (en) | 2022-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3266257B1 (en) | System and method for mobile core data services | |
| CN108282801B (en) | Switching management method based on mobile edge calculation | |
| JP5778861B2 (en) | Method and system for supporting multiple interface multiple connection communication | |
| CN109417829B (en) | Method and apparatus for determining cluster head | |
| EP3134997B1 (en) | Method for handling failure in a coordinated multicast streaming system | |
| US9357458B2 (en) | Distributed method for client optimization | |
| WO2017118186A1 (en) | Data transmission method and apparatus, and base station and ue | |
| CN109495938A (en) | Method for switching network based on multiple access edge calculations | |
| Chandrasekaran et al. | Mobility as a service (MAAS): A D2D-based information centric network architecture for edge-controlled content distribution | |
| EP3878156B1 (en) | Methods and apparatuses for content delivery over mobile networks with multi-access edge computing (mec) control and user plane separation (cups) | |
| CN104994549B (en) | A kind of communication means, access network elements and user equipment | |
| EP3135071B1 (en) | User anchor controller communication within a network environment | |
| Hagos | The performance of network-controlled mobile data offloading from LTE to WiFi networks | |
| CN113825211A (en) | Method for integrating dedicated channel signaling and service channel associated signaling | |
| CN114090221B (en) | Dual migration method for tasks in edge computing environment | |
| KR20150068038A (en) | Mobile telecommunication terminal having multiple communication interfaces and its cooperative data communication method | |
| US20180007596A1 (en) | Methods, apparatuses and computer programs for publishing updated context information pertaining to mobile terminal through publishing/subscribe mechanism | |
| CN115278865A (en) | Positioning configuration method and electronic equipment | |
| AU2021274166B2 (en) | 5G multicast broadcast service handover | |
| US20230008158A1 (en) | Systems and methods for device-assisted seamless transfer between edge computing systems in a wireless network | |
| WO2022180849A1 (en) | Network operating system and network control method | |
| JP6843833B2 (en) | Node messaging between SFNs | |
| CN116033499A (en) | Wireless access network switching method based on terminal position prediction in 5G private network environment | |
| KR20230055389A (en) | Method and apparatus for communicating in a base station using a plurality of transmission and receiption points | |
| CN118104370A (en) | Communication method and device |
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 |