CN108616401B - Intelligent video content server deployment method and system - Google Patents
Intelligent video content server deployment method and system Download PDFInfo
- Publication number
- CN108616401B CN108616401B CN201810437355.5A CN201810437355A CN108616401B CN 108616401 B CN108616401 B CN 108616401B CN 201810437355 A CN201810437355 A CN 201810437355A CN 108616401 B CN108616401 B CN 108616401B
- Authority
- CN
- China
- Prior art keywords
- video content
- network topology
- content server
- topology model
- deployment
- 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
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/12—Discovery or management of network topologies
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
Abstract
The invention provides an intelligent video content server deployment method and system, wherein the method comprises the following steps: acquiring network topology information; encoding chromosomes of the evolutionary algorithm according to the network topology information, wherein each chromosome array represents a feasible solution; then, generating a network topology model of the multiple source points and the multiple sinks according to the feasible solution, and using the network topology model as an initial server deployment scheme; converting the network topology model of the multi-source and multi-sink into the network topology model of the single-source and single-sink; determining the fitness value of each network topology model in the evolutionary algorithm through a minimum cost maximum flow algorithm; and (3) performing multiple iterations by adopting an evolutionary algorithm, continuously evolving, reserving the individual with the minimum adaptive value in each generation, and finally obtaining the optimal scheme for deploying the video content server. The invention realizes the intelligent deployment of the video content server and ensures that the deployment cost and the bandwidth leasing fee of the video content server are minimized on the premise of meeting the bandwidth requirement of a user.
Description
Technical Field
The invention relates to the fields of computer science and technology, artificial intelligence and server deployment, in particular to a method and a system for deploying a video content server by means of an artificial intelligence algorithm.
Background
With the advent of the large video age, the importance of the video viewing experience is increasingly highlighted. And the deployment location of the video content server determines the end user's viewing experience and the cost of the video service provider. Aiming at the bandwidth requirement of a user, the current technical research is mainly satisfied by reasonably distributing the bandwidth on the basis of the existing server. It has the problems that: due to the rapid increase of the number of users and the rapid increase of bandwidth requirements, the existing server cannot meet the bandwidth requirements of the users, so that the user requirements cannot be met by allocating the bandwidth.
In the research of server deployment, the prior art ignores the problems of server deployment cost and bandwidth leasing cost, thereby causing the service cost of the video service provider to be extremely large. The invention aims to minimize the deployment cost and bandwidth leasing fee of the video content server on the premise of meeting the bandwidth requirement of a user by reasonably deploying the position and bandwidth allocation of the video content server.
Disclosure of Invention
The invention provides an intelligent video content server deployment method and system, which can realize intelligent deployment of a video content server by means of intelligence and comprehensiveness of a group intelligent algorithm in an artificial intelligence algorithm when solving a multi-target problem, and can minimize the deployment cost and bandwidth leasing fee of the video content server on the premise of meeting the bandwidth requirement of a user.
The deployment method is realized by adopting the following technical scheme: an intelligent video content server deployment method comprises the following steps:
and 5, performing multiple iterations by adopting an evolutionary algorithm, continuously evolving, reserving the individual with the minimum adaptive value in each generation, and finally obtaining the optimal scheme for deploying the video content server.
Preferably, in step 2, one chromosome array corresponds to one initial server deployment scenario, and each chromosome array is randomly generated.
Preferably, in step 3, the network topology model conversion includes the following steps:
(1) establishing a super source point S and a super sink point T;
(2) for each user node ID, establishing an ID → T edge, wherein the cost is 0, and the upper limit of the capacity is the demand of the user node;
(3) for each server node NID, an S → NID edge is established with a cost of 0 and an upper capacity limit of the server node.
Preferably, in step 4, the fitness function of the evolutionary algorithm uses a minimum cost max flow algorithm, the maximum flow value of the feasible solution is used to determine whether the feasible solution can meet the requirements of all users, and the minimum cost value is used as the fitness value of the feasible solution.
Preferably, in the iterative process of step 5, the selection operation of the optimal individual keeping strategy is used; calculating the cross probability Pc of each individual, and randomly crossing according to the probability; calculating the variation probability Pm of each individual, and randomly varying according to the probability; calculating the fitness value of each chromosome of the new population; and when the iteration termination condition is met, obtaining the optimal video content server deployment scheme.
The deployment system of the invention is realized by adopting the following technical scheme: an intelligent video content server deployment system, comprising:
the information acquisition module is used for acquiring network topology information;
the video content server deployment module based on the evolutionary algorithm is used for: encoding chromosomes of the evolutionary algorithm according to the network topology information, wherein each chromosome array represents a feasible solution; then generating a network topology model of multiple source points and multiple sinks according to the feasible solution, and using the network topology model as an initial server deployment scheme; converting the network topology model of the multi-source and multi-sink into the network topology model of the single-source and single-sink; determining the value of the fitness of each network topology model in the evolutionary algorithm through a minimum cost maximum flow algorithm, namely a minimum cost value; and (3) performing multiple iterations by adopting an evolutionary algorithm, continuously evolving, reserving the individual with the minimum adaptive value in each generation, and finally obtaining the optimal scheme for deploying the video content server.
Preferably, the intelligent video content server deployment system further comprises: and the deployment result visualization module is used for visually checking the obtained optimal deployment scheme of the video content server.
Preferably, in the video content server deployment module based on the evolutionary algorithm, a process of converting the network topology model of the multi-source and multi-sink into the network topology model of the single-source and single-sink is as follows:
(1) establishing a super source point S and a super sink point T;
(2) for each user node ID, establishing an ID → T edge, wherein the cost is 0, and the upper limit of the capacity is the demand of the user node;
(3) for each server node NID, establishing an S → NID edge with the cost of 0 and the upper limit of the capacity of the server node;
the process of multiple iterations with the evolutionary algorithm is as follows:
a selection operation using an optimal individual keeping policy; calculating the cross probability Pc of each individual, and randomly crossing according to the probability; calculating the variation probability Pm of each individual, and randomly varying according to the probability; calculating the fitness value of each chromosome of the new population; and when the iteration termination condition is met, obtaining the optimal video content server deployment scheme.
According to the technical scheme for solving the deployment problem of the video content server, the feasible solution is randomly generated through a group intelligent algorithm, the deployment position of the video content server is determined, the bandwidth size distributed by each server is determined by means of a minimum cost maximum stream algorithm, and the deployment cost and the bandwidth leasing fee of the server are calculated. Because the minimum cost maximum flow algorithm can only solve the problems of a single source point and a single sink point, and the server deployment problem is the problem of multi-source point and multi-sink point, the invention comprises a model conversion method for converting a multi-source point and multi-sink server deployment model into a single-source point and single-sink point model. Compared with the prior art, the invention has the following technical effects:
(1) the intelligent deployment of the video content server is realized by means of a computer, the workload of server deployment personnel is reduced, meanwhile, the computer is higher in calculation speed, the deployment influence factors are considered more comprehensively, and the deployment result is more scientific.
(2) The deployment scheme provided by the invention adopts a group intelligent algorithm, and by reasonably deploying the position and bandwidth allocation of the video content server, the server deployment cost and bandwidth leasing fee of a video content service provider are fully considered on the premise of meeting the user requirements, so that the rationality and feasibility of the deployment result are ensured, and the total cost of the video content service provider is greatly reduced.
(3) The video content server deployment problem is an NP difficult problem, the swarm intelligence algorithm is proved to have good robustness and comprehensiveness when the NP difficult problem is solved, and the swarm intelligence algorithm is adopted to deploy the video content server, so that the accuracy of a final server deployment result can be ensured.
Drawings
FIG. 1 is a network topology diagram;
FIG. 2 is a flow chart of video content server deployment in an embodiment of the present invention;
FIG. 3 is a flow chart of an evolutionary algorithm in an embodiment of the present invention;
FIG. 4 is a schematic diagram of model conversion in an embodiment of the invention;
fig. 5 is a deployment result visualization diagram.
Detailed Description
The technical solution of the present invention will be further described with reference to the following examples and drawings, but the embodiments of the present invention are not limited thereto.
The deployment method adopts a group intelligent algorithm, and can form an intelligent video content server deployment scheme with the minimum video content server deployment cost and bandwidth leasing fee on the premise of meeting the bandwidth requirement of a user by reasonably deploying the position and bandwidth allocation of the video content server; the deployment condition of the video content server on each node is simulated by reasonably coding the individual in the algorithm, the bandwidth flowing condition of the deployment scheme of the video content server is simulated by reasonably modifying the fitness function of the algorithm, and the deployment cost and the bandwidth renting fee of the server are calculated.
The network topology of the video content service is shown in fig. 1, the numbers in the nodes are the labels of the nodes, the nodes drawn by dotted lines represent user nodes, and the other nodes are network nodes. Using need on links directly connected to user nodesjRepresenting the video bandwidth consumption requirement of the jth user, the rest nodes except the user node can be used for deploying the video content server. Using cap on each linkiRepresents the link capacity upper limit, cost, of the ith linkiIndicating the lease cost per bandwidth for the ith link. The invention aims to solve the problem that k nodes are selected to deploy a video content server in a network topological graph with m nodes, n users and e links, and the video content server meets the requirement of link bandwidth flowi<capiConditions of (2) bandwidth ∑ flow to flow from video content server to userjMeet the user's demand (i.e. ∑ flow)j≥needj) Meanwhile, the total cost (namely the sum of the server deployment cost and the bandwidth leasing fee) of deploying the video content servers is made to be (k) deployment costs plus bandwidth real costs (∑ costs)i*flowi) And minimum.
The embodiment provides an intelligent video content server deployment scheme based on an evolutionary algorithm, which is more efficient under the conditions that the number of existing network nodes is large and the bandwidth demand of users is increasing rapidly, and the intelligent management of the server deployment scheme is realized by reasonably planning the deployment position and the bandwidth flow direction of the video content server by means of a swarm intelligent algorithm, so that the deployment management efficiency is further improved.
As shown in FIG. 2, the scheme is mainly realized by an information acquisition module, a video content server deployment module based on an evolutionary algorithm and a deployment result visualization module, wherein the information acquisition module acquires required data in an existing network topological graph, inputs the acquired data into the video content server deployment module based on the evolutionary algorithm, and finally obtains a server deployment result which can meet the requirements of all users and simultaneously minimize the cost of a video content service provider, and the deployment result visualization module uses software such as MAT L AB to visually output the server deployment result, and the specific implementation steps comprise the following steps:
The step is obtained by an information obtaining module.
And 2, encoding chromosomes of the evolutionary algorithm according to the network topology information extracted in the step 1, wherein each chromosome array represents a feasible solution as shown in fig. 3, and the feasible solution is a randomly generated possible solution, namely N0 s or 1 s are randomly generated. Assuming a total of N network nodes, the ith chromosome array can be represented as:
Xi=(xi,1,xi,2,...,xi,N)
if xi,jIf it is 1, it means that a video content server is deployed on the jth network node; if xi,j0 means that no video content server is deployed on the jth network node.
And then, generating a network topology model of the multi-source point and the multi-sink point according to the feasible solution, namely randomly generating an initial population as an initial server deployment scheme. A chromosome array corresponds to an initial server deployment scenario, such as: now there are 5 network nodes on which servers can be deployed, then a randomly generated array of chromosomes (i.e., a deployment scenario) might be 01001, meaning that servers are deployed at nodes 1, 4, respectively, and no servers are deployed at nodes 0, 2, 3. While 0 and 1 are random, so a chromosome array is said to be a deployment plan, and each chromosome array is randomly generated. The number of chromosome arrays can be determined according to actual conditions, and 30 chromosome arrays are preferred in this embodiment.
And 3, converting the generated network topology model, and converting the network topology model of the multi-source point and the multi-sink point into a network topology model of a single-source point and a single-sink point.
As shown in fig. 4, the specific steps of the model conversion method are as follows:
(1) establishing a super source point S and a super sink point T;
(2) for each user node ID, establishing an ID → T edge, wherein the cost is 0, and the upper limit of the capacity is the demand of the user node;
(3) for each server node NID, an S → NID edge is established with a cost of 0 and an upper capacity limit of the server node.
After the corresponding edges are established in the steps (2) and (3), the conversion of the model is completed, and then the minimum cost and the maximum flow are determined according to the converted model.
And 4, determining the value of the fitness of each network topology model in the evolutionary algorithm through a minimum cost maximum flow algorithm, namely a minimum cost value.
After the model conversion in step 3 is completed, both the user node and the server node can be regarded as common network nodes, and the minimum cost value and the maximum flow value of S → T are calculated. And the fitness function of the evolutionary algorithm uses a minimum cost maximum flow algorithm, wherein the maximum flow value of the feasible solution is used for judging whether the feasible solution can meet the requirements of all users, and the minimum cost value is used as the adaptive value of the feasible solution.
And 5, performing multiple iterations by adopting an evolutionary algorithm, continuously evolving, reserving the individual with the minimum adaptive value in each generation, and finally obtaining the optimal scheme for deploying the video content server, wherein the optimal scheme comprises specific nodes deployed by the video content server and specific bandwidth size and direction allocated to each link.
In the iterative process of the step 5, the selection operation of the optimal individual maintenance strategy is used; calculating the cross probability Pc of each individual, and randomly crossing according to the probability; calculating the variation probability Pm of each individual, and randomly varying according to the probability; calculating the fitness value of each chromosome of the new population; and when the iteration termination condition is met, obtaining the optimal video content server deployment scheme. The iteration termination condition of this embodiment is iteration 300 generations.
The steps 2-5 are executed in the video content server deployment module based on the evolutionary algorithm.
And 6, carrying out visual inspection on the obtained optimal scheme for deploying the video content server, as shown in fig. 5, so that the video content server is easier to actually deploy.
The visualization check of this step is performed by a deployment result visualization module, such as implemented by software like MAT L AB.
As described above, the present invention can be preferably realized.
Claims (8)
1. An intelligent video content server deployment method is characterized by comprising the following steps:
step 1, acquiring network topology information;
step 2, encoding chromosomes of the evolutionary algorithm according to the network topology information acquired in the step 1, wherein each chromosome array represents a feasible solution; then, generating a network topology model of the multiple source points and the multiple sinks according to the feasible solution, and using the network topology model as an initial server deployment scheme;
step 3, converting the network topology model of the multi-source and multi-sink into the network topology model of the single-source and single-sink;
step 4, determining the value of the fitness of each network topology model in the evolutionary algorithm through a minimum cost maximum flow algorithm, namely a minimum cost value;
step 5, performing multiple iterations by adopting an evolutionary algorithm, continuously evolving, reserving the individual with the minimum adaptive value in each generation, and finally obtaining the optimal scheme for deploying the video content server;
using need on links directly connected to user nodesjRepresenting the video bandwidth consumption requirement of the jth user, wherein the other nodes except the user node can be used for deploying a video content server; each linkTop application capiRepresents the link capacity upper limit, cost, of the ith linkiIndicating the lease cost per unit bandwidth of the ith link;
in step 3, the network topology model conversion includes the following steps:
(1) establishing a super source point S and a super sink point T;
(2) for each user node ID, establishing an ID → T edge, wherein the cost is 0, and the upper limit of the capacity is the demand of the user node;
(3) for each server node NID, establishing an S → NID edge with the cost of 0 and the upper limit of the capacity of the server node;
and 3, after the network topology model conversion in the step 3 is completed, regarding the user node and the server node as common network nodes, and solving the minimum cost value and the maximum flow value of S → T.
2. The intelligent video content server deployment method according to claim 1, wherein in step 2, one chromosome array corresponds to one initial server deployment scenario, and each chromosome array is randomly generated.
3. The intelligent video content server deployment method according to claim 1, wherein the network topology information includes a number of network nodes, a number of network links, a number of users, link specific information, and user information.
4. The intelligent video content server deployment method according to claim 1, wherein in step 4, the fitness function of the evolutionary algorithm uses a minimum cost max flow algorithm, the maximum flow value of the feasible solution is used to determine whether the feasible solution can meet all user requirements, and the minimum cost value is used as the fitness value of the feasible solution.
5. The intelligent video content server deployment method according to claim 1, wherein in the iterative process of step 5, the selection operation of the optimal individual retention strategy is used; calculating the cross probability Pc of each individual, and randomly crossing according to the probability; calculating the variation probability Pm of each individual, and randomly varying according to the probability; calculating the fitness value of each chromosome of the new population; and when the iteration termination condition is met, obtaining the optimal video content server deployment scheme.
6. The intelligent video content server deployment method of claim 1, further comprising: and 6, carrying out visual inspection on the obtained optimal scheme for deploying the video content server.
7. An intelligent video content server deployment system, comprising:
the information acquisition module is used for acquiring network topology information;
the video content server deployment module based on the evolutionary algorithm is used for: encoding chromosomes of the evolutionary algorithm according to the network topology information, wherein each chromosome array represents a feasible solution; then generating a network topology model of multiple source points and multiple sinks according to the feasible solution, and using the network topology model as an initial server deployment scheme; converting the network topology model of the multi-source and multi-sink into the network topology model of the single-source and single-sink; determining the value of the fitness of each network topology model in the evolutionary algorithm through a minimum cost maximum flow algorithm, namely a minimum cost value; performing multiple iterations by adopting an evolutionary algorithm, continuously evolving, reserving the individual with the minimum adaptive value in each generation, and finally obtaining the optimal scheme for deploying the video content server;
using need on links directly connected to user nodesjRepresenting the video bandwidth consumption requirement of the jth user, wherein the other nodes except the user node can be used for deploying a video content server; using cap on each linkiRepresents the link capacity upper limit, cost, of the ith linkiIndicating the lease cost per unit bandwidth of the ith link;
in the video content server deployment module based on the evolutionary algorithm, the process of converting the network topology model of the multi-source and multi-sink into the network topology model of the single-source and single-sink is as follows:
(1) establishing a super source point S and a super sink point T;
(2) for each user node ID, establishing an ID → T edge, wherein the cost is 0, and the upper limit of the capacity is the demand of the user node;
(3) for each server node NID, establishing an S → NID edge with the cost of 0 and the upper limit of the capacity of the server node;
after the conversion of the network topology model is completed, both the user node and the server node are regarded as common network nodes, and the minimum cost value and the maximum flow value of S → T are solved;
the process of multiple iterations with the evolutionary algorithm is as follows:
a selection operation using an optimal individual keeping policy; calculating the cross probability Pc of each individual, and randomly crossing according to the probability; calculating the variation probability Pm of each individual, and randomly varying according to the probability; calculating the fitness value of each chromosome of the new population; and when the iteration termination condition is met, obtaining the optimal video content server deployment scheme.
8. The intelligent video content server deployment system of claim 7, further comprising:
and the deployment result visualization module is used for visually checking the obtained optimal deployment scheme of the video content server.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810437355.5A CN108616401B (en) | 2018-05-09 | 2018-05-09 | Intelligent video content server deployment method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810437355.5A CN108616401B (en) | 2018-05-09 | 2018-05-09 | Intelligent video content server deployment method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108616401A CN108616401A (en) | 2018-10-02 |
CN108616401B true CN108616401B (en) | 2020-07-28 |
Family
ID=63662453
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810437355.5A Active CN108616401B (en) | 2018-05-09 | 2018-05-09 | Intelligent video content server deployment method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108616401B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110198344A (en) * | 2019-05-05 | 2019-09-03 | 网宿科技股份有限公司 | A kind of resource regulating method and system |
CN110516302B (en) * | 2019-07-22 | 2022-11-29 | 新奥数能科技有限公司 | Regional intelligent energy network configuration method and device based on difference evolution algorithm |
CN110677306B (en) * | 2019-10-25 | 2021-09-03 | 上海交通大学 | Network topology replica server configuration method and device, storage medium and terminal |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103313263A (en) * | 2013-04-25 | 2013-09-18 | 中山大学 | Wireless sensor network node hierarchical scheduling method based on genetic algorithm |
CN104166630A (en) * | 2014-08-06 | 2014-11-26 | 哈尔滨工程大学 | Method oriented to prediction-based optimal cache placement in content central network |
CN107623595A (en) * | 2017-09-05 | 2018-01-23 | 湘潭大学 | A kind of webserver dispositions method and system |
CN107766941A (en) * | 2017-10-31 | 2018-03-06 | 天津大学 | A kind of facility site selecting method based on genetic algorithm |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9473354B2 (en) * | 2010-09-07 | 2016-10-18 | Bae Systems Plc | Assigning resources to resource-utilising entities |
-
2018
- 2018-05-09 CN CN201810437355.5A patent/CN108616401B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103313263A (en) * | 2013-04-25 | 2013-09-18 | 中山大学 | Wireless sensor network node hierarchical scheduling method based on genetic algorithm |
CN104166630A (en) * | 2014-08-06 | 2014-11-26 | 哈尔滨工程大学 | Method oriented to prediction-based optimal cache placement in content central network |
CN107623595A (en) * | 2017-09-05 | 2018-01-23 | 湘潭大学 | A kind of webserver dispositions method and system |
CN107766941A (en) * | 2017-10-31 | 2018-03-06 | 天津大学 | A kind of facility site selecting method based on genetic algorithm |
Non-Patent Citations (3)
Title |
---|
An Evolution Algorithm with Double-Level Archives for Multiobjective Optimization;Ni Chen etc.;《IEEE Transactions on Cybernetics》;20141016;第45卷(第9期);第1851-1863页 * |
云存储部署优化的进化算法设计;李皓等;《东南大学学报(自然科学版)》;20130924;第43卷(第z1期);第202-205页 * |
遗传算法中自适应方法的比较和分析;龚月姣等;《计算机工程与设计》;20100111;第30卷(第21期);第4907-4913页 * |
Also Published As
Publication number | Publication date |
---|---|
CN108616401A (en) | 2018-10-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yan et al. | Automatic virtual network embedding: A deep reinforcement learning approach with graph convolutional networks | |
CN108616401B (en) | Intelligent video content server deployment method and system | |
CN107330056B (en) | Wind power plant SCADA system based on big data cloud computing platform and operation method thereof | |
CN106326585B (en) | Prediction analysis method and device based on Bayesian Network Inference | |
Hao et al. | Deep reinforcement learning for edge service placement in softwarized industrial cyber-physical system | |
CN104616205A (en) | Distributed log analysis based operation state monitoring method of power system | |
CN106547882A (en) | A kind of real-time processing method and system of big data of marketing in intelligent grid | |
Xie et al. | Virtualized network function forwarding graph placing in SDN and NFV-enabled IoT networks: A graph neural network assisted deep reinforcement learning method | |
CN110430068A (en) | A kind of Feature Engineering method of combination and device | |
CN109743286A (en) | A kind of IP type mark method and apparatus based on figure convolutional neural networks | |
CN106597968A (en) | Converter high-speed real-time monitoring system and method based on Redis | |
CN112506691A (en) | Method and system for recovering digital twin application fault of multi-energy system | |
CN104618480A (en) | Cloud system source distributing method driven on basis of network link utilization rates | |
Fajjari et al. | Cloud networking: An overview of virtual network embedding strategies | |
CN107911763B (en) | Intelligent power distribution and utilization communication network EPON network planning method based on QoS | |
CN109688068A (en) | Network load balancing method and device based on big data analysis | |
US9124496B2 (en) | System and method for end- or service-node placement optimization | |
CN112464545B (en) | Layout method, system, equipment and medium for cables and transformer substation of offshore wind farm | |
Kavitha et al. | Dynamic load balancing in cloud based multimedia system with genetic algorithm | |
CN115310538A (en) | Microgrid green electricity mark tracing method and device | |
CN111294553B (en) | Method, device, equipment and storage medium for processing video monitoring service signaling | |
CN112165721A (en) | Multi-service task unloading and service migration method based on edge computing | |
CN112698944A (en) | Distributed cloud computing system and method based on human brain simulation | |
CN115314394B (en) | Resource allocation method for smart power grid | |
Shahin | Memetic elitist Pareto evolutionary algorithm for virtual network embedding |
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 |