CN114666409A - Service migration method based on cache management in edge computing environment - Google Patents
Service migration method based on cache management in edge computing environment Download PDFInfo
- Publication number
- CN114666409A CN114666409A CN202210180888.6A CN202210180888A CN114666409A CN 114666409 A CN114666409 A CN 114666409A CN 202210180888 A CN202210180888 A CN 202210180888A CN 114666409 A CN114666409 A CN 114666409A
- Authority
- CN
- China
- Prior art keywords
- edge
- base station
- service
- cloud center
- analyzed
- 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
-
- 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
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1021—Server selection for load balancing based on client or server locations
-
- 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
- H04W28/10—Flow control between communication endpoints
- H04W28/14—Flow control between communication endpoints using intermediate storage
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention relates to a service migration method based on cache management in an edge computing environment, which divides resources on edge nodes into three parts including a running state service occupied resource, a halt state service occupied resource and an idle resource, monitors the resources required by all service operation on the edge nodes, analyzes the edge cloud center of a base station according to service to be migrated, executes running state service analysis, halt state service analysis, edge node idle resource analysis and halt state service occupied resource release analysis layer by layer so as to complete a service migration design strategy, introduces a time attenuation strategy therein, provides a sequential basis for the execution of halt state service resource release, integrates the whole scheme design, improves the hit rate of service cache in the edge nodes, and can effectively reduce the migration times while completing basic service migration, and maximally utilizes the caching function of the edge node.
Description
Technical Field
The invention relates to a service migration method based on cache management in an edge computing environment, and belongs to the technical field of edge computing.
Background
With the popularization of various intelligent mobile devices, a mobile terminal generates many new applications with high computing intensity or large storage requirements, but the mobile device is limited by computing power and storage space and cannot meet the requirements of the new applications. Meanwhile, if the computing task and the storage task of the mobile application are both handed to a cloud computing server, on one hand, the burden of the cloud server is increased, and on the other hand, the application response delay is large due to network congestion and transmission delay, and the service quality is reduced.
Edge computing and related techniques have evolved to overcome the resource limitations of mobile devices and to provide low-latency, high-quality services to users. Edge computing allows the deployment of the computing resources and application environments of a cloud server to the edge of a cellular network, which is made up of several edge nodes with certain computing power, storage resources and network resources. The edge calculation is a new network architecture formed by fusing the internet technology and the communication technology, and the basic idea behind the edge calculation is that the application and the related program are operated at a place closer to a user, so that the network transmission delay can be effectively reduced, and the network congestion can be effectively reduced. Edge computing is completed by relying on edge nodes, the edge nodes are physically a single small server with complete functions, the computing capacity and other resources of the edge nodes are far lower than those of a cloud server, and therefore the edge nodes are generally configured with a light-weight module to rapidly configure an operating environment and deploy various services.
In an edge computing environment, a cloud center needs to deploy a part of computing and storage resources to an edge node closer to a mobile terminal, and run a service required by a mobile terminal application on the edge node, so that network transmission delay can be effectively reduced and network congestion can be reduced. However, in the actual operation process of the service, the mobile terminal may move with various distances along with the user, which may cause the mobile terminal to gradually move away from the edge node originally providing the service for the mobile terminal in the moving process, and the transmission delay of the service on the edge node and the terminal interaction may increase accordingly. With the increase of the transmission delay between the service on the primary edge node and the terminal, the primary edge node cannot provide low-delay and high-quality interaction for the user, and at this time, the service on the primary edge node needs to be migrated to an edge node closer to the user, but the main difficulty of service migration is the selection of a migration strategy, namely how to select an optimal edge node near the user to place the service to meet the user requirement.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a service migration method based on cache management in an edge computing environment, design a brand new logic strategy, quickly adjust the position of a corresponding service according to the mobile state of a mobile terminal, improve high-efficiency computing processing for the mobile terminal and improve the working efficiency of edge computing.
The invention adopts the following technical scheme for solving the technical problems: the invention designs a service migration method based on cache management in an edge computing environment, which is based on each base station edge cloud center divided according to a base station radiation range, each edge node in a radiation range respectively corresponding to each base station edge cloud center and each service respectively loaded by each edge node, and executes the following steps aiming at a target mobile terminal after moving from a position in the radiation range corresponding to a base station edge cloud center P to a position in the radiation range corresponding to a base station edge cloud center Q, wherein the position in the radiation range corresponding to the base station edge cloud center P is the position in the edge node corresponding to the base station edge cloud center P, and the target mobile terminal is provided with a target service for computing processing:
step A, obtaining the distance between each base station edge cloud center and a target mobile terminal respectively, sequencing the base station edge cloud centers according to the sequence of the distance from small to large, namely, the first base station edge cloud center is a base station edge cloud center Q, selecting the first base station edge cloud center to a base station edge cloud center P in sequence to serve as each base station edge cloud center to be analyzed, maintaining the sequencing between the base station edge cloud centers to be analyzed, initializing n to 1, and entering the step B;
b, judging whether a target service in an operation state exists in each service loaded on each edge node corresponding to the edge cloud center of the nth base station to be analyzed, if so, randomly selecting one target service in the operation state, and providing calculation processing for the target mobile terminal through the edge cloud center of the nth base station to be analyzed; otherwise, entering the step C;
step C, judging whether target services exist in the shutdown state services loaded on each edge node corresponding to the nth base station edge cloud center to be analyzed, if yes, randomly starting one of the shutdown state target services, and providing computing processing for the target mobile terminal through the nth base station edge cloud center to be analyzed; otherwise, entering the step D;
step D, judging whether edge nodes with preset idle quantity of each target type resource meeting the demand quantity of each target type resource corresponding to the target service exist in each edge node corresponding to the nth base station edge cloud center to be analyzed, if so, randomly selecting one edge node from each edge node meeting the condition, loading and starting the target service on the edge node, and providing calculation processing for the target mobile terminal through the nth base station edge cloud center to be analyzed; otherwise, entering the step E;
step E, randomly selecting one edge node which does not participate in the processing from the step F to the step I from all edge nodes corresponding to the edge cloud center of the nth base station to be analyzed as an edge node to be analyzed, and entering the step F;
f, sequencing the shutdown state services loaded on the edge node to be analyzed according to the shutdown state duration corresponding to the shutdown state services from large to small, initializing k to 1, and entering the step G;
step G, respectively aiming at each target type resource, obtaining the sum of the occupied amount of the target type resource respectively corresponding to the 1 st shutdown state service to the kth shutdown state service in the sequence and the sum of the idle amount of the target type resource corresponding to the edge node to be analyzed, further obtaining the sum corresponding to each target type resource, judging whether the sum corresponding to each target type resource meets the demand of the target service corresponding to each target type resource, and if so, entering the step H; otherwise, entering step I;
step H, releasing the 1 st halt state service to the kth halt state service, wherein the occupied amount of the target type resources is respectively corresponding to idle amount, deleting the 1 st halt state service to the kth halt state service from the edge node to be analyzed, then loading and starting the target service on the edge node to be analyzed, and providing calculation processing for the target mobile terminal through the nth base station edge cloud center to be analyzed;
step I, judging whether k is equal to the total number of the loaded shutdown state services of the edge node to be analyzed, if so, entering step J; otherwise, updating by adding 1 according to the value of k, and returning to the step G;
step J, judging whether edge nodes which do not participate in the processing from step F to step I exist in edge nodes corresponding to an nth base station edge cloud center to be analyzed, if so, returning to step E, otherwise, entering step K;
step K, judging whether the (n + 1) th base station edge cloud center to be analyzed is a base station edge cloud center P, if so, providing computing processing for the target mobile terminal by the running state target service loaded on the edge node corresponding to the base station edge cloud center P and continuously passing through the base station edge cloud center P; otherwise, updating by adding 1 for the value of n, and returning to the step B.
As a preferred technical scheme of the invention: step i, in step B and step C, respectively randomly selecting a target service in an operating state, randomly starting a target service in a shutdown state, providing computing processing for the target mobile terminal through the nth base station edge cloud center to be analyzed, and then entering step i; in the step D, randomly selecting one edge node from all edge nodes meeting the condition, loading and starting a target service on the edge node, providing calculation processing for a target mobile terminal through an nth base station edge cloud center to be analyzed, and then entering the step i; and after step H is executed, then step i is executed;
step i, judging whether a target service in an operation state in a base station edge cloud center P provides calculation processing for the mobile terminal or not, and if so, not performing any further processing; otherwise, the target service in the running state is stopped, so that the target service is in a shutdown state.
As a preferred technical scheme of the invention: in the step D, based on the edge nodes corresponding to the nth base station edge cloud center to be analyzed, which have the edge nodes with the preset idle amount of each target type resource that all satisfy the demand amount of the target service corresponding to each target type resource, the edge node corresponding to the specified target type resource in each target type resource with the maximum idle amount is selected, the target service is loaded and started on the edge node, and the target mobile terminal is provided with computing processing through the nth base station edge cloud center to be analyzed.
As a preferred technical scheme of the invention: if not, initializing l to 1, and entering the step E;
in the step E, based on all edge nodes corresponding to the n-th base station edge cloud center to be analyzed and according to the sequence of the idle quantity of the designated target type resources in all target type resources corresponding to the n-th base station edge cloud center from large to small, the l-th edge node is selected as the edge node to be analyzed, and the step F is carried out;
in the step J, judging whether l is equal to the number of edge nodes corresponding to the edge cloud center of the nth base station to be analyzed, if so, entering the step K; otherwise, adding 1 to update for the value of l, and returning to the step E.
As a preferred technical scheme of the invention: in the step F, for each shutdown state service loaded on the edge node to be analyzed, the following formula is used:
priorityn,l,k=2*endn,l,k-startn,l,k-cur_time
obtaining the shutdown state duration corresponding to each shutdown state service, and then carrying out the operation according to the shutdown state duration corresponding to the shutdown state service from large to smallSorting, then initializing k to 1, and entering step G; wherein the priorityn,l,kRepresenting the shutdown state duration corresponding to the loaded kth shutdown state service on the ith edge node corresponding to the nth base station edge cloud center to be analyzed, cur _ time representing the current time, endn,l,kThe stopping time of the latest stopping at the distance cur _ time corresponding to the loaded kth stopping state service on the ith edge node corresponding to the edge cloud center of the nth base station to be analyzed is represented as startn,l,kThe starting time of the latest starting of the loaded kth shutdown state service on the ith edge node corresponding to the edge cloud center of the nth base station to be analyzed from the cur _ time is represented.
As a preferred technical scheme of the invention: the preset target type resources comprise RAM occupation amount, MIPS demand amount and bandwidth demand amount.
Compared with the prior art, the service migration method based on cache management in the edge computing environment has the following technical effects by adopting the technical scheme:
(1) the invention designs a service migration method based on cache management in an edge computing environment, which divides resources on edge nodes into three parts, including resources occupied by a running state service, resources occupied by a halt state service and idle resources, monitors the resources required by the running of all services on the edge nodes, analyzes aiming at a base station edge cloud center according to a service to be migrated, executes running state service analysis, halt state service analysis, edge node idle resource analysis and halt state service occupied resource release analysis layer by layer, further completes a service migration design strategy, introduces a time attenuation strategy, provides a sequence basis for the execution of the halt state service resource release, integrates the whole scheme design, improves the hit rate of a service cache in the edge nodes, and can effectively reduce the migration times while completing basic service migration, and maximally utilizes the caching function of the edge node.
Drawings
FIG. 1 is a flow chart illustrating a method for service migration based on cache management in an edge computing environment according to the present invention.
Detailed Description
The following description will explain embodiments of the present invention in further detail with reference to the accompanying drawings.
The invention designs a service migration method based on cache management in an edge computing environment, which is based on each base station edge cloud center divided according to a base station radiation range, each edge node in a radiation range respectively corresponding to each base station edge cloud center and each service respectively loaded by each edge node, such as a base station edge cloud center set D ═ D1,d2,…,d|D|D, wherein the edge cloud center of the nth base station in the set D is DnAnd D | represents the number of base station edge cloud centers in the set D, and defines the n-th base station edge cloud center DnEdge node set with different physical resourcesDefining the n base station edge cloud center dnMiddle ith edge node en,lLoaded service aggregation
And for each service, designing the specific design of each target type resource, wherein each target type resource comprises RAM (unit MB) occupation amount, MIPS (unit million instruction per second) demand amount and bandwidth (unit bit/s) demand amount, and based on each preset target type resource, combining the specific combination of the services loaded by the edge node, including running state service and shutdown state service, three sets can be specifically constructed, wherein a running state service resource use condition set R is definedn,lAs follows, | Rn,lI represents the n base station edge cloud center dnMiddle ith edge node en,lThe number of the upper running state services, r _ r represents the occupation amount of the running state service RAM (unit MB), m _ r represents the demand amount of the running state service MIPS (unit million instructions per second), and bw _ r represents the demand amount of the running state service bandwidth (unit bit/s).
Defining a set H of outage state service resource usagen,lAs follows, | Hn,lI represents the n base station edge cloud center dnMiddle ith edge node en,lThe number of the up-down state services, r _ h represents the occupation amount of the down-state service RAM (unit MB), m _ h represents the demand amount of the down-state service MIPS (unit million instructions per second), and bw _ h represents the demand amount of the down-state service bandwidth (unit bit/s).
Defining a set of free resources Fn,lR _ f as followsn,lRepresenting the n-th base station edge cloud center dnMiddle ith edge node en,lUpper RAM (unit MB) free amount, m _ fn,lRepresenting the n-th base station edge cloud center dnMiddle ith edge node en,lUpper MIPS (units of million instructions per second) idle, bw _ fn,lRepresenting the n-th base station edge cloud center dnMiddle ith edge node en,lUpper bandwidth (unit bit/s) idle amount.
Fn,l=[r_fn,l m_fn,l bw_fn,l]
Correspondingly, an Active operation state service list is constructed for each edge node respectivelyn,lService list Sleep of the halt staten,lAnd recording its start time start for each service on the edge noden,l,vEnd of stop timen,l,vAnd the current clock time cur _ time is acquired.
In practical application, as shown in fig. 1, the following steps are specifically executed after a target mobile terminal moves from a position in a radiation range corresponding to a base station edge cloud center P to a position in a radiation range corresponding to a base station edge cloud center Q, with respect to a load on an edge node corresponding to the base station edge cloud center P and a target service for providing calculation processing for the target mobile terminal.
And step A, obtaining the distance between each base station edge cloud center and a target mobile terminal, sequencing the base station edge cloud centers according to the sequence from small to large of the distance, namely, a first base station edge cloud center is a base station edge cloud center Q, selecting the first base station edge cloud center to a base station edge cloud center P in sequence as each base station edge cloud center to be analyzed, keeping the sequencing between the base station edge cloud centers to be analyzed, initializing n to 1, and entering the step B.
B, judging whether a target service in an operation state exists in each service loaded on each edge node corresponding to the edge cloud center of the nth base station to be analyzed, if so, randomly selecting one target service in the operation state, providing calculation processing for the target mobile terminal through the edge cloud center of the nth base station to be analyzed, and then entering the step L; otherwise, entering the step C.
Step C, judging whether target services exist in the shutdown state services loaded on each edge node corresponding to the nth base station edge cloud center to be analyzed, if yes, randomly starting one of the shutdown state target services, providing calculation processing for the target mobile terminal through the nth base station edge cloud center to be analyzed, and combining an Active list of running state services on the edge node in practical applicationn,lService list Sleep of the halt staten,lThe shutdown state is serviced by Sleepn,lDelete and add Activen,lThen, go to step L; otherwise, entering the step D.
D, judging whether edge nodes with preset idle amount of each target type resource meeting the demand of each target type resource corresponding to the target service exist in each edge node corresponding to the nth base station edge cloud center to be analyzed, if so, randomly selecting one edge node from each edge node meeting the following formula (1) of the condition, loading and starting the target service on the edge node, providing calculation processing for the target mobile terminal through the nth base station edge cloud center to be analyzed, and entering the step L; otherwise, initializing l to 1, and entering step E.
Wherein, r _ rTarget taskM _ r, representing the RAM demand for the target taskTarget taskIndicating the MIPS demand corresponding to the target task, bw _ rTarget taskAnd representing the bandwidth demand corresponding to the target task.
In the specific implementation of step D, based on the idle amount of the resource in each edge node corresponding to the edge cloud center of the base station to be analyzed, it is determined whether there is an edge node corresponding to the edge cloud center of the nth base station to be analyzed, where the idle amount of each preset target type resource meets the demand of the target service for each target type resource, and specifically, if yes, the edge node corresponding to the specified target type resource in each target type resource is selected, the target service is loaded and started on the edge node, and the target mobile terminal is provided with calculation processing through the edge cloud center of the nth base station to be analyzed, that is, the target service is added into the running state service list Activen,lPerforming the following steps; otherwise, initializing l to 1, and entering step E.
And E, selecting the l-th edge node as the edge node to be analyzed based on the edge nodes corresponding to the n-th base station edge cloud center to be analyzed and the sequencing of the idle quantity of the designated target type resources in the target type resources corresponding to the n-th base station edge cloud center from large to small, and entering the step F.
And F, sequencing the shutdown state services loaded on the edge node to be analyzed according to the shutdown state duration corresponding to the shutdown state services from large to small, initializing the state duration k to 1, and entering the step G.
In practical application, in the step F, for each shutdown state service loaded on the edge node to be analyzed, the following formula (2) is used:
priorityn,l,k=2*endn,l,k-startn,l,k-cur_time (2)
obtaining the time length of the shutdown state corresponding to each shutdown state serviceThen sorting the shutdown state duration corresponding to the shutdown state service from large to small, initializing k to 1, and entering a step G; wherein the priorityn,l,kRepresenting the shutdown state duration corresponding to the loaded kth shutdown state service on the ith edge node corresponding to the nth base station edge cloud center to be analyzed, cur _ time representing the current time, endn,l,kRepresenting the shutdown time, start, of the latest shutdown at the distance cur _ time corresponding to the loaded kth shutdown state service on the ith edge node corresponding to the edge cloud center of the nth base station to be analyzedn,l,kThe starting time of the latest starting of the loaded kth shutdown state service on the ith edge node corresponding to the edge cloud center of the nth base station to be analyzed from the cur _ time is represented.
Step G, respectively aiming at each target type resource, obtaining the sum of the occupied amount of the target type resource corresponding to the 1 st shutdown state service to the kth shutdown state service in the sequence and the sum of the idle amount of the target type resource corresponding to the edge node to be analyzed, further obtaining the sum corresponding to each target type resource, and judging whether the sum corresponding to each target type resource meets the demand of the target service corresponding to each target type resource, namely the following formula (3), if yes, entering the step H; otherwise, entering step I.
Step H, releasing the 1 st halt state service to the kth halt state service, wherein the occupied amount of the corresponding target type resources is idle, deleting the 1 st halt state service to the kth halt state service from the edge node to be analyzed, loading a starting target service on the edge node to be analyzed, providing calculation processing for the target mobile terminal through the nth base station edge cloud center to be analyzed, and entering step L, specifically, deleting Sleepn,lThe 1 st to the k-th shutdown state services, and releases the resources, while adding the target service to Activen,lIn (1).
Step I, judging whether k is equal to the total number of the loaded shutdown state services of the edge node to be analyzed, if so, entering step J; otherwise, updating by adding 1 for the value of k, and returning to the step G.
Step J, judging whether l is equal to the number of edge nodes corresponding to the edge cloud center of the nth base station to be analyzed, if so, entering the step K; otherwise, updating by adding 1 to the value of l, and returning to the step E.
Step K, judging whether the (n + 1) th base station edge cloud center to be analyzed is a base station edge cloud center P, if so, providing computing processing for a target mobile terminal by an operation state target service loaded on an edge node corresponding to the base station edge cloud center P and continuously passing through the base station edge cloud center P; otherwise, updating by adding 1 according to the value of n, and returning to the step B.
L, judging whether the target service in the running state in the base station edge cloud center P provides computing processing for the mobile terminal or not, and if so, not performing any further processing; otherwise, the target service in the running state is stopped and is in a shutdown state, namely the target task is deleted from the running state service list and added into the shutdown state service list.
The service migration method based on cache management in the edge computing environment designed by the technical scheme divides resources on edge nodes into three parts, including operating state service occupied resources, halt state service occupied resources and idle resources, monitors resources required by all service operations on the edge nodes, analyzes the edge cloud center of the base station according to the service to be migrated, executes operating state service analysis, halt state service analysis, edge node idle resource analysis and halt state service occupied resource release analysis layer by layer, further completes a service migration design strategy, introduces a time attenuation strategy therein, provides a sequential basis for the execution of halt state service resource release, integrates the whole scheme design, improves the hit rate of service caches in the edge nodes, and can effectively reduce the migration times while completing basic service migration, and maximally utilizes the caching function of the edge node.
The embodiments of the present invention have been described in detail with reference to the drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the gist of the present invention.
Claims (6)
1. A service migration method based on cache management in an edge computing environment is characterized in that: based on each base station edge cloud center divided according to the base station radiation range, each edge node in the radiation range corresponding to each base station edge cloud center, and each service loaded by each edge node, after the target mobile terminal moves from the position in the radiation range corresponding to the base station edge cloud center P to the position in the radiation range corresponding to the base station edge cloud center Q, the following steps are executed with respect to the load on the edge node corresponding to the base station edge cloud center P and the target service providing calculation processing for the target mobile terminal:
step A, obtaining the distance between each base station edge cloud center and a target mobile terminal respectively, sequencing the base station edge cloud centers according to the sequence of the distance from small to large, namely, the first base station edge cloud center is a base station edge cloud center Q, selecting the first base station edge cloud center to a base station edge cloud center P in sequence to serve as each base station edge cloud center to be analyzed, maintaining the sequencing between the base station edge cloud centers to be analyzed, initializing n to 1, and entering the step B;
b, judging whether a target service in an operation state exists in each service loaded on each edge node corresponding to the edge cloud center of the nth base station to be analyzed, if so, randomly selecting one target service in the operation state, and providing calculation processing for the target mobile terminal through the edge cloud center of the nth base station to be analyzed; otherwise, entering the step C;
step C, judging whether target services exist in the shutdown state services loaded on each edge node corresponding to the nth base station edge cloud center to be analyzed, if yes, randomly starting one of the shutdown state target services, and providing computing processing for the target mobile terminal through the nth base station edge cloud center to be analyzed; otherwise, entering the step D;
d, judging whether edge nodes with preset idle quantity of each target type resource meeting the demand quantity of the target service corresponding to each target type resource exist in each edge node corresponding to the nth base station edge cloud center to be analyzed, if so, randomly selecting one edge node from each edge node meeting the condition, loading and starting the target service on the edge node, and providing calculation processing for the target mobile terminal through the nth base station edge cloud center to be analyzed; otherwise, entering the step E;
step E, randomly selecting one edge node which does not participate in the processing from the step F to the step I from all edge nodes corresponding to the edge cloud center of the nth base station to be analyzed as an edge node to be analyzed, and entering the step F;
f, sequencing the shutdown state services loaded on the edge node to be analyzed according to the shutdown state duration corresponding to the shutdown state services from large to small, initializing k to 1, and entering the step G;
step G, respectively aiming at each target type resource, obtaining the sum of the occupied amount of the target type resource respectively corresponding to the 1 st shutdown state service to the kth shutdown state service in the sequence and the sum of the idle amount of the target type resource corresponding to the edge node to be analyzed, further obtaining the sum corresponding to each target type resource, judging whether the sum corresponding to each target type resource meets the demand of the target service corresponding to each target type resource, and if so, entering the step H; otherwise, entering step I;
step H, releasing the 1 st halt state service to the kth halt state service, wherein the occupied amount of the target type resources is respectively corresponding to idle amount, deleting the 1 st halt state service to the kth halt state service from the edge node to be analyzed, then loading and starting the target service on the edge node to be analyzed, and providing calculation processing for the target mobile terminal through the nth base station edge cloud center to be analyzed;
step I, judging whether k is equal to the total number of the loaded shutdown state services of the edge node to be analyzed, if so, entering step J; otherwise, updating by adding 1 according to the value of k, and returning to the step G;
step J, judging whether edge nodes which do not participate in the processing from step F to step I exist in edge nodes corresponding to the nth base station edge cloud center to be analyzed, if yes, returning to step E, and otherwise, entering step K;
step K, judging whether the (n + 1) th base station edge cloud center to be analyzed is a base station edge cloud center P, if so, providing computing processing for the target mobile terminal by the running state target service loaded on the edge node corresponding to the base station edge cloud center P and continuously passing through the base station edge cloud center P; otherwise, updating by adding 1 for the value of n, and returning to the step B.
2. The method for service migration based on cache management in an edge computing environment according to claim 1, wherein: step i, in step B and step C, respectively randomly selecting a target service in an operating state, randomly starting a target service in a shutdown state, providing computing processing for the target mobile terminal through the nth base station edge cloud center to be analyzed, and then entering step i; in the step D, randomly selecting one edge node from all edge nodes meeting the condition, loading and starting a target service on the edge node, providing calculation processing for a target mobile terminal through an nth base station edge cloud center to be analyzed, and then entering the step i; and after step H is executed, then step i is executed; step i, judging whether a target service in an operation state in a base station edge cloud center P provides calculation processing for the mobile terminal or not, and if so, not performing any further processing; otherwise, the target service in the running state is stopped, so that the target service is in a shutdown state.
3. The method for service migration based on cache management in an edge computing environment according to claim 1, wherein: in the step D, based on the edge nodes corresponding to the nth base station edge cloud center to be analyzed, which have the edge nodes with the preset idle amount of each target type resource that all satisfy the demand amount of the target service corresponding to each target type resource, the edge node corresponding to the specified target type resource in each target type resource with the maximum idle amount is selected, the target service is loaded and started on the edge node, and the target mobile terminal is provided with computing processing through the nth base station edge cloud center to be analyzed.
4. The method for service migration based on cache management in an edge computing environment according to claim 1, wherein: if not, initializing l to 1, and entering the step E;
in the step E, based on all edge nodes corresponding to the n-th base station edge cloud center to be analyzed and according to the sequence of the idle quantity of the designated target type resources in all target type resources corresponding to the n-th base station edge cloud center from large to small, the l-th edge node is selected as the edge node to be analyzed, and the step F is carried out;
in the step J, judging whether l is equal to the number of edge nodes corresponding to the edge cloud center of the nth base station to be analyzed, if so, entering the step K; otherwise, adding 1 to update for the value of l, and returning to the step E.
5. The method for service migration based on cache management in an edge computing environment according to claim 1, wherein:
in the step F, for each shutdown state service loaded on the edge node to be analyzed, the following formula is used:
priorityn,l,k=2*endn,l,k-startn,l,k-cur_time
obtaining the corresponding halt state duration of each halt state service, then sorting the halt state durations corresponding to the halt state services from large to small, then initializing k to 1, and entering the step G; wherein the priorityn,l,kThe method comprises the steps of representing the shutdown state duration corresponding to the kth shutdown state service loaded on the l edge node corresponding to the edge cloud center of the nth base station to be analyzed, cur _ time represents the current time, and endn,l,kRepresenting the shutdown time, start, of the latest shutdown at the distance cur _ time corresponding to the loaded kth shutdown state service on the ith edge node corresponding to the edge cloud center of the nth base station to be analyzedn,l,kRepresenting the edge cloud center of the nth base station to be analyzedThe starting time of the last starting from cur _ time corresponding to the kth shutdown state service loaded on the corresponding ith edge node.
6. The method for service migration based on cache management in an edge computing environment according to claim 1, wherein: the preset target type resources comprise RAM occupation amount, MIPS demand amount and bandwidth demand amount.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210180888.6A CN114666409B (en) | 2022-02-26 | 2022-02-26 | Service migration method based on cache management in edge computing environment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210180888.6A CN114666409B (en) | 2022-02-26 | 2022-02-26 | Service migration method based on cache management in edge computing environment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114666409A true CN114666409A (en) | 2022-06-24 |
CN114666409B CN114666409B (en) | 2023-07-25 |
Family
ID=82028290
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210180888.6A Active CN114666409B (en) | 2022-02-26 | 2022-02-26 | Service migration method based on cache management in edge computing environment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114666409B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117014313A (en) * | 2023-09-26 | 2023-11-07 | 工业云制造(四川)创新中心有限公司 | Method and system for analyzing equipment data of edge cloud platform in real time |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109600419A (en) * | 2018-11-12 | 2019-04-09 | 南京信息工程大学 | The computation migration method of car networking application is supported under a kind of mobile edge calculations environment |
CN109714382A (en) * | 2018-09-18 | 2019-05-03 | 贵州电网有限责任公司 | A kind of Multi-User Multi-Task migration decision-making technique of non-equilibrium edge cloud MEC system |
US20190158606A1 (en) * | 2018-12-28 | 2019-05-23 | Francesc Guim Bernat | QUALITY OF SERVICE (QoS) MANAGEMENT IN EDGE COMPUTING ENVIRONMENTS |
US20190342797A1 (en) * | 2016-12-07 | 2019-11-07 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and migration managing module for managing a migration of a service |
CN111031102A (en) * | 2019-11-25 | 2020-04-17 | 哈尔滨工业大学 | Multi-user, multi-task mobile edge computing system cacheable task migration method |
-
2022
- 2022-02-26 CN CN202210180888.6A patent/CN114666409B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190342797A1 (en) * | 2016-12-07 | 2019-11-07 | Telefonaktiebolaget Lm Ericsson (Publ) | Method and migration managing module for managing a migration of a service |
CN109714382A (en) * | 2018-09-18 | 2019-05-03 | 贵州电网有限责任公司 | A kind of Multi-User Multi-Task migration decision-making technique of non-equilibrium edge cloud MEC system |
CN109600419A (en) * | 2018-11-12 | 2019-04-09 | 南京信息工程大学 | The computation migration method of car networking application is supported under a kind of mobile edge calculations environment |
US20190158606A1 (en) * | 2018-12-28 | 2019-05-23 | Francesc Guim Bernat | QUALITY OF SERVICE (QoS) MANAGEMENT IN EDGE COMPUTING ENVIRONMENTS |
CN111031102A (en) * | 2019-11-25 | 2020-04-17 | 哈尔滨工业大学 | Multi-user, multi-task mobile edge computing system cacheable task migration method |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117014313A (en) * | 2023-09-26 | 2023-11-07 | 工业云制造(四川)创新中心有限公司 | Method and system for analyzing equipment data of edge cloud platform in real time |
CN117014313B (en) * | 2023-09-26 | 2023-12-19 | 工业云制造(四川)创新中心有限公司 | Method and system for analyzing equipment data of edge cloud platform in real time |
Also Published As
Publication number | Publication date |
---|---|
CN114666409B (en) | 2023-07-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110413392B (en) | Method for formulating single task migration strategy in mobile edge computing scene | |
CN111031102B (en) | Multi-user, multi-task mobile edge computing system cacheable task migration method | |
CN107911478B (en) | Multi-user calculation unloading method and device based on chemical reaction optimization algorithm | |
CN108958916B (en) | Workflow unloading optimization method under mobile edge environment | |
US10268469B2 (en) | Systems and methods for controlling application updates across a wireless interface | |
CN109788046B (en) | Multi-strategy edge computing resource scheduling method based on improved bee colony algorithm | |
JP2002269023A5 (en) | ||
CN110069341B (en) | Method for scheduling tasks with dependency relationship configured according to needs by combining functions in edge computing | |
WO2019050952A1 (en) | Systems, methods, and media for distributing database queries across a metered virtual network | |
CN112291335B (en) | Optimized task scheduling method in mobile edge calculation | |
CN115454589A (en) | Task scheduling method and device and Kubernetes scheduler | |
CN113822456A (en) | Service combination optimization deployment method based on deep reinforcement learning in cloud and mist mixed environment | |
CN116069512B (en) | Serverless efficient resource allocation method and system based on reinforcement learning | |
Chen et al. | Latency minimization for mobile edge computing networks | |
CN113422812B (en) | Service chain deployment method and device | |
CN111767146A (en) | Distributed machine learning system acceleration method based on network reconfiguration | |
CN114666409A (en) | Service migration method based on cache management in edge computing environment | |
CN114938381A (en) | D2D-MEC unloading method based on deep reinforcement learning and computer program product | |
CN110048966B (en) | Coflow scheduling method for minimizing system overhead based on deadline | |
CN114691372A (en) | Group intelligent control method of multimedia end edge cloud system | |
CN116302578B (en) | QoS (quality of service) constraint stream application delay ensuring method and system | |
CN116302507A (en) | Application service dynamic deployment and update method based on vacation queuing | |
CN109474664A (en) | A kind of active pre-cache method and device in heterogeneous wireless network | |
CN115185660A (en) | Unloading and buffer storage method and system for MAR task in multi-access edge calculation | |
CN108628540A (en) | Data storage device and method |
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 |