CN112600827B - Virtual service migration method and system based on incremental minimum cost maximum flow - Google Patents
Virtual service migration method and system based on incremental minimum cost maximum flow Download PDFInfo
- Publication number
- CN112600827B CN112600827B CN202011435918.0A CN202011435918A CN112600827B CN 112600827 B CN112600827 B CN 112600827B CN 202011435918 A CN202011435918 A CN 202011435918A CN 112600827 B CN112600827 B CN 112600827B
- Authority
- CN
- China
- Prior art keywords
- service
- node
- edge server
- edge
- maximum flow
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
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
- H04L67/131—Protocols for games, networked simulations or virtual reality
-
- 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
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
- H04L67/63—Routing a service request depending on the request content or context
Abstract
The invention relates to the technical field of information, and discloses a virtual service migration method and a system based on incremental minimum cost maximum flow, wherein the method comprises the following steps: constructing a network topological graph according to the geographic position and the connection relation of the edge server; calculating the shortest path distance between each pair of edge servers and generating a shortest path distance matrix; establishing a minimum cost maximum flow model by combining the access information of each service, wherein the model comprises a service node and an edge server node; solving the model to obtain the minimum cost maximum flow, and transferring and placing the service nodes; updating the access information of each service, and updating the position of a service node by adopting an incremental minimum cost maximum flow algorithm; and carrying out real-time monitoring and statistics, and adjusting the corresponding service nodes and edge server nodes when the change rate of the access information exceeds a set threshold value. The invention can solve the optimal service placement and migration scheme in a short time, and is simple, reliable and easy to realize.
Description
Technical Field
The present invention relates to the field of information technologies, and in particular, to a virtual service migration method and system based on incremental minimum cost maximum flow.
Background
In recent years, Mobile Edge Computing (MEC) has become a trend, and is a cloud computing platform that deploys servers on nodes such as routers, gateways, and base stations closer to the periphery of users and centers on cloud computing technology. For some delay-sensitive applications, such as applications of online services like facial recognition, real-time games, live video, etc., the traditional cloud computing center cannot well meet the delay requirements of these applications. However, with MEC, the cloud center pushes part of the computing resources to the edge of the network, bringing some services closer to the user. The method can relieve the load pressure of the cloud computing center, reduce the service delay to a certain extent and reduce the network flow, thereby better meeting the requirements of delay-sensitive application. Although the advantages of MEC are clear, it tends to cause an increase in service delay if the service is fixed on one edge server, considering user mobility, temporal diversity of access to the service and limited coverage of the individual edge servers. Therefore, if these factors are not taken into account, the service provided may significantly increase access delay, and worse still, a large amount of network traffic may be generated, causing network congestion, resulting in a severe degradation of service quality, or even service interruption.
Currently, dynamic programming algorithms and machine learning methods are often used to solve this problem. Learners propose an adaptive configuration algorithm for virtual machine execution based on a reinforcement learning algorithm so as to solve the problem of workload of a data center. Also, some researchers have proposed a service migration mechanism that can migrate computing services to available servers for adaptive grid computing. The scholars also define a cost model aiming at the migration problem and provide a migration framework which has sensing capability, is more flexible and has strong adaptability, but aims at a single service.
Existing work has achieved good effects on migration strategies, but with the increasing demand of people, the network scale also increases, the access mode is more complex, and the previous strategies are difficult to make optimal decisions under a large-scale network topology structure. Meanwhile, the prior method aims at the migration of single service, or lacks certain dynamic property, or has overlong solving time.
Disclosure of Invention
The invention aims to provide a virtual service migration method and a virtual service migration system based on incremental minimum cost maximum flow aiming at the technical problems in the prior art, which can solve an optimal service placement and migration scheme in a short time, are simple, reliable and easy to realize.
In order to solve the problems proposed above, the technical scheme adopted by the invention is as follows:
the invention provides a virtual service migration method based on incremental minimum cost maximum flow, which comprises the following steps:
constructing a network topological graph according to the geographic position and the connection relation of the edge server;
calculating the shortest path distance between each pair of edge servers and generating a shortest path distance matrix;
establishing a minimum cost maximum flow model according to the distance matrix and by combining the access information of each service, wherein the model comprises service nodes and edge server nodes;
solving the model to obtain the minimum cost maximum flow, migrating the service nodes and placing the service nodes on the corresponding edge server nodes;
updating the access information of each service, and updating the position of a service node by adopting an incremental minimum cost maximum flow algorithm;
and monitoring and counting the access information of each service in real time, and adjusting the corresponding service node and the edge server node when the change rate of the access information exceeds a set threshold value.
Further, the minimum cost maximum flow model specifically includes:
the source node S is only connected with the service node Ni, each service node Ni is connected with all edge server nodes Ej, and each edge server node Ej is connected with the end node T; each edge in the model relates capacity and communication cost.
Further, the updating the location of the service node by using the incremental minimum cost maximum flow algorithm specifically includes:
updating the access cost and the migration cost of each service node placed on the edge server node in a residual network of the model solving algorithm;
according to the statistical access information, carrying out corresponding deletion and addition on service nodes and edge service nodes in the residual network;
and solving by using an incremental minimum cost maximum flow algorithm, and updating the position of the service node on the edge server node according to the minimum cost maximum flow obtained by solving.
Further, the incremental minimum cost maximum flow algorithm specifically includes:
the access amount of the service node is changed, and when the variable of the access times is larger than the threshold value obtained by pre-learning, the position of the service node is updated;
when the number of the service nodes is changed, deleting the service nodes or adding the service nodes, and updating the positions of the service nodes;
the edge server load capacity changes and the location of the service node is updated when the capacity of the edge server node is reduced and the outgoing flow from the edge server node is greater than the reduced capacity of the edge or the edge server node is deleted, and when the capacity of the edge server node is increased or a new edge server node is added.
Further, when the access volume of the service node changes, the method specifically includes:
when the variable of the access times is larger than the threshold value obtained by pre-learning, modifying the weight value of the side connected with the service node in the last residual network;
adding migration cost on the edge of the service node connected to the edge server node, and searching the minimum cost maximum flow again from the service node with changed cost;
and updating the position of the service node according to the obtained minimum cost maximum flow.
Further, when the number of the service nodes changes, the method specifically includes:
when the number of service nodes is reduced, deleting the service nodes and the related edges in the residual network;
when a service node is increased, adding the service node and a related edge in a residual network, selecting a minimum cost path from the service node to an edge server node, and adding migration cost on the edge of the service node connected to the edge server node;
and searching a new minimum cost maximum flow from the deleted or added service nodes, and updating the positions of the service nodes.
Further, when the load capacity of the edge server changes, the method specifically includes:
when the capacity of the edge server is reduced and the outflow stream is larger than the reduced capacity of the corresponding edge or the edge server node is deleted, returning the stream corresponding to the edge with the maximum cost; starting from the corresponding edge service node, selecting the shortest path to the end node, adding the flow, and adding the migration cost on the edge of the service node connected to the edge server node; starting from the corresponding service node, searching a new minimum cost maximum flow, and updating the position of the service node;
when the capacity of the edge server is increased or a new edge server node is added, modifying the residual network, and adding migration cost on the edge of the service node connected to the edge server node; and starting from the corresponding edge server node, searching a new minimum cost maximum flow, and updating the position of the service node.
The invention also provides a virtual service migration system based on incremental minimum cost maximum flow, which comprises the following steps:
the topological graph building module: the system comprises a network topology graph and a network topology graph, wherein the network topology graph is used for constructing the network topology graph according to the geographical position and the connection relation of an edge server;
a matrix generation module: the shortest path distance matrix is used for calculating the shortest path distance between each pair of edge servers and generating a shortest path distance matrix;
a model building module: the system comprises a distance matrix, a minimum cost maximum flow model and an edge server node, wherein the distance matrix is used for establishing the minimum cost maximum flow model according to the distance matrix and by combining the access information of each service;
the service node migration module: the system comprises a model, an edge server node and a service node, wherein the model is used for solving to obtain a minimum cost maximum flow, migrating the service node and placing the service node on the corresponding edge server node;
the service node updating module: the access information used for updating each service, and the position of the service node is updated by adopting an incremental minimum cost maximum flow algorithm;
a monitoring and adjusting module: the method is used for monitoring and counting the access information of each service in real time, and when the change rate of the access information exceeds a set threshold value, the corresponding service node and edge node are adjusted.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of any of the methods when executing the program.
The invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of any of the methods.
Compared with the prior art, the invention has the beneficial effects that:
the method and the system provided by the invention convert the communication cost problem of multiple access services in the edge network into the problem of minimum cost and maximum flow, and compared with a dynamic programming algorithm and a heuristic algorithm, the time complexity is reduced, and the total access cost is greatly reduced; and an incremental minimum cost maximum flow algorithm is adopted, an optimal service placement and migration scheme which minimizes the overall service access cost can be solved in a short time, sudden release quantity increase and abnormal conditions of the service access quantity can be well responded, the method is more efficient than the existing dynamic planning method, greedy method and machine learning method, the method is simple, reliable and easy to implement, quick response to users is guaranteed, and the service quality is improved.
Drawings
In order to illustrate the solution of the invention more clearly, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are some embodiments of the invention, and that other drawings may be derived from these drawings by a person skilled in the art without inventive effort. Wherein:
fig. 1 is a flowchart illustrating a virtual service migration method according to the present invention.
FIG. 2 is a network topology diagram of virtual service migration in accordance with the present invention.
FIG. 3 is a diagram of a minimum cost maximum flow model according to the present invention.
FIG. 4 is a flow chart of the incremental minimum cost max flow algorithm of the present invention.
Fig. 5 is a schematic diagram of the last residual network of the present invention.
Fig. 6 is a schematic diagram of a modified residual network of the present invention.
FIG. 7 is a schematic diagram of a virtual service migration system of the present invention.
Fig. 8 is a schematic diagram of an electronic device according to the present invention.
Detailed Description
To facilitate an understanding of the invention, the invention will now be described more fully with reference to the accompanying drawings. Preferred embodiments of the present invention are shown in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.
Referring to fig. 1, the present invention provides a virtual service migration method based on incremental minimum cost maximum flow, which includes the following specific steps:
and step S1, constructing a network topological graph according to the geographical position and the connection relation of the edge server. Specifically, the weight of the edge in the network topology is determined by the distance between the connected edge nodes and the network transmission speed.
The application scene of the invention can be abstracted into a network topological graph as shown in fig. 2, because the virtual service runs on the server of the edge node in the edge network, the edge node is abstracted into the edge server node, namely the edge servers 1-5 connected with the decision center in fig. 2, and the user can access the corresponding edge server to the service required by the user through the nearest access point 1-5.
And step S2, calculating the shortest path distance between each pair of edge servers in the network topological graph and generating a shortest path distance matrix. Specifically, the shortest path distance is calculated by using a shortest path algorithm, and the method is simple and reliable.
Step S3: and counting the access information of each service in the nearest time period, and establishing a minimum cost maximum flow model according to the shortest path distance matrix, wherein the model comprises service nodes and edge server nodes. Specifically, according to the shortest path distance matrix, the access cost and the migration cost of each service placed on each edge node are calculated, and a minimum cost maximum flow model is established.
In the embodiment of the present invention, referring to fig. 3, the process of establishing the minimum cost maximum flow model specifically includes:
and taking the S node and the T node as a source node and a destination node of the model, which are set for constructing the model. Ni represents the ith service node, Ej represents the jth edge server node, wherein i is 1-n, j is 1-m, namely, n service nodes and m edge server nodes are shared. The source node S is connected with only the service node Ni, each service node Ni is connected with all the edge server nodes Ej, and each edge server node Ej is connected with the terminal node T. Each edge in the model is associated with two values, capacity and communication cost respectively.
Further, the capacity of each edge connecting the source node S and the service node Ni is 1, which enables all services to be placed at the edge server or the cloud center, and the cost is 0. The capacity of the edge connecting each service node Ni with all edge server nodes Ej is 1, and the cost is the sum of the migration cost and the delay cost of placing the service node on the edge server node. The capacity of the edge connecting each edge server node Ej and the terminal node T is the number of services that the edge server node can bear, and the cost is 0.
Step S4: and solving the model to obtain the minimum cost maximum flow, migrating the service nodes and placing the service nodes on the corresponding edge server nodes.
Specifically, a classical minimum cost maximum flow solving algorithm is used for solving, if a flow flows from a service node Ni to an edge server node Ej in the solved minimum cost maximum flow, the service node Ni is placed on the edge server node Ej, the service node is migrated and placed according to the flow, namely the edge server node corresponding to the service node can be obtained according to a solving result, and then the service node is migrated and placed. In the embodiment of the invention, the classical minimum cost maximum flow solving algorithm adopts a continuous shortest path method, a scaling method or a circle elimination method, and can reliably and effectively obtain the minimum cost maximum flow.
Step S5: and updating the access information of each service, namely counting the access information of each service again at intervals, and updating the position of the corresponding service node on the edge server node by adopting an incremental minimum cost maximum flow algorithm.
Specifically, in the residual network of the model solving algorithm, the access cost and the migration cost of each service node placed on each edge server node are updated; performing corresponding deletion and addition on service nodes and edge server nodes in the residual network according to the counted access information; and performing fast solving by using an incremental minimum cost maximum flow algorithm, and updating the position of the service node on the edge server node according to the minimum cost maximum flow obtained by solving.
In the embodiment of the invention, as the access mode of the user to the service changes frequently, the access information of the service can be counted and updated again at intervals. According to the conventional method, the steps S3 and S4 are selected to be carried out again, but the solving time is too long. The embodiment of the invention directly modifies the corresponding service node and the corresponding service edge in the residual network which is used for solving the algorithm last time by utilizing the incremental minimum cost maximum flow algorithm, but not constructing a minimum cost maximum flow model from the beginning and solving, and executing a ring elimination method on the modified residual network, thereby reducing the solving time by 8-15 times.
Step S6: and monitoring and counting the access information of each service in real time, and adjusting the corresponding service node and edge node when the change rate of the access information exceeds a set threshold value.
Specifically, the number of the users accessing the service nodes may have sudden dramatic increase in the amount of the access and abnormal conditions, and the monitor monitors and counts the access information in real time. When the change rate of the monitored or counted access information exceeds the set threshold, the residual network is updated immediately, i.e., the incremental minimum cost maximum flow algorithm proposed in step S5 is used to solve the problem, and the service node and the edge server node are adjusted accordingly.
In the embodiment of the present invention, the position of each service node is updated by using an incremental minimum cost maximum flow algorithm, as shown in fig. 4, which specifically includes the following steps:
step S51: the method for updating the position of the service node when the access amount of the user to the service node is changed and the set condition is met includes:
step S511: when the variable of the access times is larger than the threshold value obtained by pre-learning, modifying the weight value of the side connected with the service node in the last residual network; otherwise, the weight value is not modified, so that the number of edges needing to be modified is reduced.
In the embodiment of the invention, considering that the access quantity of most service nodes changes at intervals, and the access times of users to some service nodes can not change too much, the access times can be compared with the optimal threshold value obtained by learning in advance through a machine learning method to determine whether the weight of the edge needs to be modified. Furthermore, a threshold value of the associated edge weight value of the service node to be modified is determined by using a machine learning method, the threshold value is not set by using experience or a traditional method, but an optimal threshold value is obtained by using the machine learning method and previous data set training, and the obtained threshold value can adapt to various access modes, large-scale high-frequency changing network structures and various abnormal conditions of a user, so that a decision center cannot frequently change the residual network structures, the service adjustment times are reduced, and the service quality is improved.
Step S512: and adding migration cost on the edge of the service node connected to the edge server node, and searching the minimum cost maximum flow again from the service node with changed cost. Specifically, the minimum cost maximum flow is searched again by using the circle elimination method, and the optimal scheme can be found out faster than other schemes by using the circle elimination method because the minimum cost maximum flow is sent on the basis of the minimum cost maximum flow in the last stage.
Step S513: and updating the position of the service node on the edge server node according to the obtained minimum cost maximum flow.
In the embodiment of the present invention, assuming that the last adjusted partial residual network is shown in fig. 5, the capacity and weight of the edge from the service node N1 to the edge server node E1 are (0, 100), which indicates that the service node N1 has been placed on the edge server node E1, and generates the corresponding reverse edge (1, -100). The service node Nn is distributed to the edge server node E2 in the same way. After a period of time, the number of accesses to the service nodes N1 and Nn by the user changes, and the number of accesses is greater than the threshold, the corresponding edge in the residual network needs to be modified, and the modified residual network is shown in fig. 6.
As can be seen from fig. 5, the negative loop of N1 → E2 → Nn → E1 → N1 is formed, and the loop elimination method can be used to eliminate the loop from the service node N1 until no negative loop exists in the residual network, at which time the maximum streaming minimum cost is reached.
Step S52: when the number of the service nodes changes, deleting the service nodes or adding the service nodes, and updating the positions of the service nodes, specifically comprising:
when the number of service nodes is reduced, deleting the service nodes and the related edges in the residual network;
when a service node is increased, adding the service node and related edges in the residual network, selecting a minimum cost path from the service node to an edge server node, and adding a flow to reach a maximum flow, thereby realizing the placement of all service nodes; adding migration cost on the edge of the service node connected to the edge server node;
and starting from the deleted or added service nodes and flowing to the edge server node, searching a new flow with the minimum cost and the maximum cost, and migrating and placing the service nodes to complete updating. Specifically, a new minimum cost maximum stream is also searched by using a circle elimination method.
Step S53: when the load capacity of the edge server changes, the position of the service node is updated, which specifically comprises:
step S531: when the capacity of the edge server is reduced, checking whether the flow flowing out from the edge server node is larger than the reduced capacity of the corresponding edge, specifically:
if the outgoing flow is still smaller than the reduced capacity of the edge, only changing the capacity of the edge in the residual network, and not adjusting the residual network;
if the outgoing flow is larger than the reduced capacity of the edge or the deletion of the edge server node occurs, returning the flow corresponding to the edge with the highest cost flowing into the edge server node; starting from the corresponding edge service node, selecting the shortest path to the end node, adding the flow, and adding the migration cost on the edge of the service node connected to the edge server node;
starting from the corresponding service node, searching a new minimum cost maximum flow by using a circle elimination method, and then migrating and placing the service node to complete the updating of the position of the service node;
step S532: when the capacity of the edge server is increased or a new edge server node is added, modifying the residual network, and adding migration cost on the edge of the service node connected to the edge server node; and starting from the corresponding edge server node, searching a new minimum cost maximum flow by using a circle elimination method, and updating the position of the service node.
The virtual service migration method provided by the embodiment of the invention builds the minimum cost maximum flow model by building the network topological graph and generating the shortest path distance matrix. And solving to obtain the minimum cost maximum flow, transferring and placing the service nodes, and updating the corresponding service nodes by adopting an incremental minimum cost maximum flow algorithm at intervals. And monitoring and counting the access information of each service in real time, and adjusting the corresponding service node and edge node as required. The embodiment of the invention converts the total delay problem of multiple service accesses in the edge network into the problem of minimum cost maximum flow, obtains the optimal placement and migration strategy for minimizing the overall service access cost by adopting an incremental minimum cost maximum flow algorithm, can solve the strategy in a short time, and can well deal with the sudden release amount increase and abnormal conditions of the service access quantity.
Referring to fig. 7, an embodiment of the present invention further provides a virtual service migration system based on incremental minimum cost maximum flow, where the system includes the following:
the topological graph building module: and the method is used for constructing the network topology map according to the geographical position and the connection relation of the edge server.
A matrix generation module: for calculating the shortest path distance between each pair of edge servers and generating a shortest path distance matrix.
A model building module: and the system is used for establishing a minimum cost maximum flow model according to the distance matrix and by combining the access information of each service, wherein the model comprises service nodes and edge server nodes.
The service node migration module: and the method is used for solving the model to obtain the minimum cost maximum flow, migrating the service nodes and placing the service nodes on the corresponding edge server nodes.
The service node updating module: and updating the position of the service node by adopting an incremental minimum cost maximum flow algorithm.
A monitoring and adjusting module: the method is used for monitoring and counting the access information of each service in real time, and when the change rate of the access information exceeds a set threshold value, the corresponding service node and the edge server node are adjusted.
Specifically, the system provided in the embodiment of the present invention is specifically configured to execute the method embodiment described above, and details of the method embodiment of the present invention are not described again.
Fig. 8 is a schematic entity structure diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 8, the electronic device may include: a processor (processor)301, a communication Interface (communication Interface)302, a memory (memory)303 and a communication bus 304, wherein the processor 301, the communication Interface 302 and the memory 303 complete communication with each other through the communication bus 304. The processor 301 may invoke a computer program stored on the memory 303 and executable on the processor 301 to perform the methods provided by the various embodiments described above, including, for example:
constructing a network topological graph according to the geographic position and the connection relation of the edge server;
calculating the shortest path distance between each pair of edge servers and generating a shortest path distance matrix;
establishing a minimum cost maximum flow model according to the distance matrix and by combining the access information of each service, wherein the model comprises service nodes and edge server nodes;
solving the model to obtain the minimum cost maximum flow, migrating the service nodes and placing the service nodes on the corresponding edge server nodes;
updating the access information of each service, and updating the position of a service node by adopting an incremental minimum cost maximum flow algorithm;
and monitoring and counting the access information of each service in real time, and adjusting the corresponding service node and the edge server node when the change rate of the access information exceeds a set threshold value.
In addition, the logic instructions in the memory 303 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or make a contribution to the prior art, or may be implemented in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Embodiments of the present invention further provide a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program is implemented to perform the method provided in the foregoing embodiments when executed by a processor, and the method includes:
constructing a network topological graph according to the geographic position and the connection relation of the edge server;
calculating the shortest path distance between each pair of edge servers and generating a shortest path distance matrix;
establishing a minimum cost maximum flow model according to the distance matrix and by combining the access information of each service, wherein the model comprises service nodes and edge server nodes;
solving the model to obtain the minimum cost maximum flow, migrating the service nodes and placing the service nodes on the corresponding edge server nodes;
updating the access information of each service, and updating the position of a service node by adopting an incremental minimum cost maximum flow algorithm;
and monitoring and counting the access information of each service in real time, and adjusting the corresponding service node and the edge server node when the change rate of the access information exceeds a set threshold value.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
The virtual service migration method and system based on the incremental minimum cost maximum flow solve the problem of placement and migration decision of multiple services in the edge network and improve the condition of minimum service access total delay. In a small scale network, it can calculate the placement and migration scheme of the service node within tens of milliseconds. For a large-scale network, if the number of service requests suddenly increases and the frequency of the service requests irregularly changes, the placing and transferring strategies can be solved within tens of seconds by using the incremental algorithm, so that the problem of low efficiency of the conventional service transferring algorithm is solved, the delay of the user for accessing the service can be greatly reduced, and the service quality and the user experience are improved.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (10)
1. A virtual service migration method based on incremental minimum cost maximum flow is characterized in that: the method comprises the following steps:
constructing a network topological graph according to the geographic position and the connection relation of the edge server;
calculating the shortest path distance between each pair of edge servers and generating a shortest path distance matrix;
establishing a minimum cost maximum flow model according to the distance matrix and by combining the access information of each service, wherein the model comprises service nodes and edge server nodes;
solving the model to obtain the minimum cost maximum flow, migrating the service nodes and placing the service nodes on the corresponding edge server nodes;
updating the access information of each service, and updating the position of a service node by adopting an incremental minimum cost maximum flow algorithm;
and monitoring and counting the access information of each service in real time, and adjusting the corresponding service node and the edge server node when the change rate of the access information exceeds a set threshold value.
2. The incremental minimum-cost maximum-flow-based virtual service migration method according to claim 1, wherein: the minimum cost maximum flow model specifically comprises:
the source node S is only connected with the service node Ni, each service node Ni is connected with all edge server nodes Ej, and each edge server node Ej is connected with the end node T; each edge in the model relates capacity and communication cost.
3. The incremental minimum-cost maximum-flow-based virtual service migration method according to claim 2, wherein: the updating of the position of the service node by using the incremental minimum cost maximum flow algorithm specifically comprises:
updating the access cost and the migration cost of each service node placed on the edge server node in a residual network of the model solving algorithm;
according to the statistical access information, carrying out corresponding deletion and addition on service nodes and edge service nodes in the residual network;
and solving by using an incremental minimum cost maximum flow algorithm, and updating the position of the service node on the edge server node according to the minimum cost maximum flow obtained by solving.
4. The incremental minimum-cost maximum-flow-based virtual service migration method according to claim 3, wherein: the incremental minimum cost maximum flow algorithm specifically comprises:
the access amount of the service node is changed, and when the variable of the access times is larger than the threshold value obtained by pre-learning, the position of the service node is updated;
when the number of the service nodes is changed, deleting the service nodes or adding the service nodes, and updating the positions of the service nodes;
the edge server load capacity changes and the location of the service node is updated when the capacity of the edge server node is reduced and the outgoing flow from the edge server node is greater than the reduced capacity of the corresponding edge or the edge server node is deleted, and when the capacity of the edge server node is increased or a new edge server node is added.
5. The incremental minimum-cost maximum-flow-based virtual service migration method according to claim 4, wherein: when the access volume of the service node changes, the method specifically includes:
when the variable of the access times is larger than the threshold value obtained by pre-learning, modifying the weight value of the side connected with the service node in the last residual network;
adding migration cost on the edge of the service node connected to the edge server node, and searching the minimum cost maximum flow again from the service node with changed cost;
and updating the position of the service node according to the obtained minimum cost maximum flow.
6. The incremental minimum-cost maximum-flow-based virtual service migration method according to claim 4, wherein: when the number of the service nodes changes, the method specifically includes:
when the number of service nodes is reduced, deleting the service nodes and the related edges in the residual network;
when a service node is increased, adding the service node and a related edge in a residual network, selecting a minimum cost path from the service node to an edge server node, and adding migration cost on the edge of the service node connected to the edge server node;
and searching a new minimum cost maximum flow from the reduced or increased service nodes, and updating the positions of the service nodes.
7. The incremental minimum-cost maximum-flow-based virtual service migration method according to claim 4, wherein: when the load capacity of the edge server changes, the method specifically includes:
when the capacity of the edge server is reduced and the outflow stream is larger than the reduced capacity of the corresponding edge or the edge server node is deleted, returning the stream corresponding to the edge with the maximum cost; starting from the corresponding edge service node, selecting the shortest path to the end node, adding the flow, and adding the migration cost on the edge of the service node connected to the edge server node; starting from the corresponding service node, searching a new minimum cost maximum flow, and updating the position of the service node;
when the capacity of the edge server is increased or a new edge server node is added, modifying the residual network, and adding migration cost on the edge of the service node connected to the edge server node; and starting from the corresponding edge server node, searching a new minimum cost maximum flow, and updating the position of the service node.
8. A virtual service migration system based on incremental minimum cost maximum flow, the system comprising:
the topological graph building module: the system comprises a network topology graph and a network topology graph, wherein the network topology graph is used for constructing the network topology graph according to the geographical position and the connection relation of an edge server;
a matrix generation module: the shortest path distance matrix is used for calculating the shortest path distance between each pair of edge servers and generating a shortest path distance matrix;
a model building module: the system comprises a distance matrix, a minimum cost maximum flow model and an edge server node, wherein the distance matrix is used for establishing the minimum cost maximum flow model according to the distance matrix and by combining the access information of each service;
the service node migration module: the system comprises a model, an edge server node and a service node, wherein the model is used for solving to obtain a minimum cost maximum flow, migrating the service node and placing the service node on the corresponding edge server node;
the service node updating module: the access information used for updating each service, and the position of the service node is updated by adopting an incremental minimum cost maximum flow algorithm;
a monitoring and adjusting module: the method is used for monitoring and counting the access information of each service in real time, and when the change rate of the access information exceeds a set threshold value, the corresponding service node and edge node are adjusted.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method according to any of claims 1 to 7 are implemented when the processor executes the program.
10. A non-transitory computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011435918.0A CN112600827B (en) | 2020-12-10 | 2020-12-10 | Virtual service migration method and system based on incremental minimum cost maximum flow |
| PCT/CN2020/138720 WO2022120953A1 (en) | 2020-12-10 | 2020-12-23 | Virtual service migration method and system based on incremental minimum-cost maximum-flow |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011435918.0A CN112600827B (en) | 2020-12-10 | 2020-12-10 | Virtual service migration method and system based on incremental minimum cost maximum flow |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112600827A CN112600827A (en) | 2021-04-02 |
| CN112600827B true CN112600827B (en) | 2021-10-29 |
Family
ID=75191798
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011435918.0A Active CN112600827B (en) | 2020-12-10 | 2020-12-10 | Virtual service migration method and system based on incremental minimum cost maximum flow |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112600827B (en) |
| WO (1) | WO2022120953A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114996278B (en) * | 2022-06-27 | 2023-05-23 | 华中科技大学 | Road network shortest path distance query method based on reinforcement learning |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015502616A (en) * | 2012-01-27 | 2015-01-22 | エンパイア テクノロジー ディベロップメント エルエルシー | Parameterized dynamic model for cloud migration |
| CN106708625A (en) * | 2016-12-08 | 2017-05-24 | 中国科学院软件研究所 | Minimum-cost maximum-flow based large-scale resource scheduling system and minimum-cost maximum-flow based large-scale resource scheduling method |
| CN108632813A (en) * | 2018-05-21 | 2018-10-09 | 北京邮电大学 | The motion management method and system of mobile edge calculations |
| CN111148174A (en) * | 2019-12-13 | 2020-05-12 | 北京邮电大学 | Service migration path selection method in mobile edge calculation |
| CN111158903A (en) * | 2019-12-12 | 2020-05-15 | 中国科学院深圳先进技术研究院 | Dynamic data planning method |
| CN111491332A (en) * | 2020-04-20 | 2020-08-04 | 中国石油大学(北京) | Dynamic service migration method and device |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE112018007463T5 (en) * | 2018-04-11 | 2020-12-24 | Intel IP Corporation | Flexible Multiple Access Edge Computing Service Consumption (Mec Service Consumption) through host zoning |
| CN111443990B (en) * | 2020-03-25 | 2023-04-07 | 中南大学 | Edge calculation task migration simulation system |
| CN112015518B (en) * | 2020-08-27 | 2022-11-25 | 山东大学 | Method and system for realizing real-time migration of multiple virtual machines in incremental deployment SDN environment |
-
2020
- 2020-12-10 CN CN202011435918.0A patent/CN112600827B/en active Active
- 2020-12-23 WO PCT/CN2020/138720 patent/WO2022120953A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015502616A (en) * | 2012-01-27 | 2015-01-22 | エンパイア テクノロジー ディベロップメント エルエルシー | Parameterized dynamic model for cloud migration |
| CN106708625A (en) * | 2016-12-08 | 2017-05-24 | 中国科学院软件研究所 | Minimum-cost maximum-flow based large-scale resource scheduling system and minimum-cost maximum-flow based large-scale resource scheduling method |
| CN108632813A (en) * | 2018-05-21 | 2018-10-09 | 北京邮电大学 | The motion management method and system of mobile edge calculations |
| CN111158903A (en) * | 2019-12-12 | 2020-05-15 | 中国科学院深圳先进技术研究院 | Dynamic data planning method |
| CN111148174A (en) * | 2019-12-13 | 2020-05-12 | 北京邮电大学 | Service migration path selection method in mobile edge calculation |
| CN111491332A (en) * | 2020-04-20 | 2020-08-04 | 中国石油大学(北京) | Dynamic service migration method and device |
Non-Patent Citations (1)
| Title |
|---|
| 《移动边缘计算网络中任务调度与资源配置的协同优化研究》;李雨晴;《信息科技辑》;20200615(第06期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112600827A (en) | 2021-04-02 |
| WO2022120953A1 (en) | 2022-06-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yu et al. | Computation offloading for mobile edge computing: A deep learning approach | |
| Chamola et al. | Latency aware mobile task assignment and load balancing for edge cloudlets | |
| CN109756578B (en) | Low-delay task scheduling method for dynamic fog computing network | |
| CN108076158B (en) | Minimum load route selection method and system based on naive Bayes classifier | |
| CN104038392A (en) | Method for evaluating service quality of cloud computing resources | |
| CN110809275B (en) | Micro cloud node placement method based on wireless metropolitan area network | |
| Zhang et al. | Cooperative edge caching via federated deep reinforcement learning in fog-rans | |
| CN109639833A (en) | A kind of method for scheduling task based on wireless MAN thin cloud load balancing | |
| US20220369200A1 (en) | Clustering and routing method and system for wireless sensor networks | |
| CN114465945B (en) | SDN-based identification analysis network construction method | |
| EP4024212A1 (en) | Method for scheduling interference workloads on edge network resources | |
| CN112600827B (en) | Virtual service migration method and system based on incremental minimum cost maximum flow | |
| CN111988787A (en) | Method and system for selecting network access and service placement positions of tasks | |
| CN111526036B (en) | Short flow real-time optimization method, system and network transmission terminal | |
| Yu et al. | Collaborative computation offloading for multi-access edge computing | |
| Xu et al. | An actor-critic-based transfer learning framework for experience-driven networking | |
| Chen et al. | Deep learning-based traffic prediction for energy efficiency optimization in software-defined networking | |
| Wang et al. | Reinforcement learning-based optimization for mobile edge computing scheduling game | |
| CN113361113B (en) | Energy-consumption-adjustable twin data distribution method for high-speed rail bogie | |
| Kashyap et al. | DECENT: Deep learning enabled green computation for edge centric 6G networks | |
| CN112601232B (en) | Load balancing multi-service migration method and system based on minimum cost and maximum flow | |
| CN112399483B (en) | Constraint route control system based on GNs graph predictor | |
| CN113915741B (en) | Instruction sending method, device and system | |
| Wang et al. | Evolutionary game based heterogeneous wireless network selection with multiple traffics in 5G | |
| CN107333320A (en) | Data forwarding 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 |