CN117914773A - A secure and low-latency two-layer network resource allocation method based on SDN/NFV network slicing technology - Google Patents
A secure and low-latency two-layer network resource allocation method based on SDN/NFV network slicing technology Download PDFInfo
- Publication number
- CN117914773A CN117914773A CN202311872057.6A CN202311872057A CN117914773A CN 117914773 A CN117914773 A CN 117914773A CN 202311872057 A CN202311872057 A CN 202311872057A CN 117914773 A CN117914773 A CN 117914773A
- Authority
- CN
- China
- Prior art keywords
- delay
- nis
- sdn
- resource allocation
- nfv
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/12—Shortest path evaluation
- H04L45/121—Shortest path evaluation by minimising delays
-
- 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
- H04L41/122—Discovery or management of network topologies of virtualised topologies, e.g. software-defined networks [SDN] or network function virtualisation [NFV]
-
- 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
- H04L41/147—Network analysis or design for predicting network behaviour
-
- 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
- H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a safe and low-delay double-layer network resource allocation method based on SDN/NFV network slicing technology, which comprises a double-layer network resource allocation framework based on SDN/NFV network slicing technology so as to fully utilize network resources; secondly, deploying an SDN/NFV-based power grid infrastructure environment, defining end-to-end flow delay, service processing delay and packet delivery time estimation, and restricting a short path to establish an objective function for minimizing delay; finally we improve the delay minimization of the network topology by two ways of NIS middlebox placement based on optimized kmeans-based graph cluster analysis and dynamic resource allocation based on regression tree analysis to predict NIS usage. Compared with the existing intelligent power grid technology, the safety and low-delay double-layer network resource allocation method based on the SDN/NFV network slicing technology can improve the network safety level of the power network and reduce delay, and provides a more intelligent network resource allocation scheme for the future power interconnection network.
Description
Technical Field
The invention belongs to the technical field of electric power Internet of things, and particularly relates to a safe and low-delay double-layer network resource allocation method based on SDN/NFV network slicing technology.
Background
Future 5G networks will pose considerable challenges for power internet of things requiring advanced services such as real-time high quality services, intelligent metering, etc., specifically this includes not only a substantial increase in overall data rate, but also a decrease in end-to-end delay, an increase in energy efficiency, and a large-scale connection. Mobile broadband communications is the driving force for 5G network traffic growth. Thus, the grid will need to invest in new solutions or enhance its infrastructure through radio access technologies, fronthaul, backhaul, etc. Typically, as traffic increases, the grid deploys new dedicated network devices for the control plane and the data plane. These deployments are often excessive in view of the anticipated loads during peak hours of three to five years in the future. And some network resources are often wasted, resulting in policy inefficiency in terms of capital expenditures and operational expenditures. In order to better utilize the available resources, a more flexible, agile and economical solution has arisen, which relies on the features provided by Network Function Virtualization (NFV) and Software Defined Networks (SDN).
NFV deploys network functions as Virtual Network Function (VNFs) software instances, which provides important benefits to the grid in terms of cost, flexibility, openness, configuration homogeneity, migration, etc. NFV is applicable to any data plane packet processing and control plane functions in fixed and mobile network infrastructure. NFV is currently solving the use case of mobile core virtualization, and has recently received great attention.
On the other hand, SDN may handle logically centralized control, implementing the programmability of the network by decoupling the data and control planes. SDN provides a logical plane abstraction that hides the hardware specific to a department of the power grid, thereby facilitating interoperability of departments of the power grid. This abstraction enables network virtualization, i.e., partitioning (slicing) of the physical infrastructure, to create multiple co-existence and independent network tenants thereon.
In recent years, smart grids have been supporting and approving research into embedded artificial intelligence, SDN, and Network Function Virtualization (NFV). In the network security context, with the service policy as input, a set of artificial intelligence driven Network Intelligent Services (NIS) can be virtually applied in NFV-based middleware overlaid on an SDN architecture, becoming an SDN/NFV-based network infrastructure. However, using this approach to protect the application data stream may significantly increase the end-to-end stream delay. In view of the definition of reliability requirements, successful delivery of the application data stream, but with delays higher than the defined requirements, can be considered as failure.
A balance between the required network security and low-delay communication is considered essential for many applications in the power grid. The inventors have found that in practice the end-to-end flow delay always depends on the position of the middlebox in the network. However, studies have demonstrated that this problem is a non-deterministic polynomial time problem, a complex and time consuming decision problem. Thus, there is a need for a trade-off optimization method to implement a heuristic solution that provides the effort of delay optimization NIS in SDN/NFV based grid infrastructure.
Disclosure of Invention
In order to solve the technical problems, the invention provides a safe and low-delay double-layer network resource allocation method based on SDN/NFV network slicing technology, so as to improve the network safety level of a power network and reduce delay.
The invention provides a safe and low-delay double-layer network resource allocation method based on SDN/NFV network slicing technology, which comprises a double-layer network resource allocation framework based on SDN/NFV network slicing technology so as to fully utilize network resources; secondly, deploying an SDN/NFV-based power grid infrastructure environment, defining end-to-end flow delay, service processing delay and packet delivery time estimation, and restricting a short path to establish an objective function for minimizing delay; finally we improve the delay minimization of the network topology by two ways of NIS middlebox placement based on optimized kmeans-based graph cluster analysis and dynamic resource allocation based on regression tree analysis to predict NIS usage. Compared with the existing intelligent power grid technology, the safety and low-delay double-layer network resource allocation method based on the SDN/NFV network slicing technology can improve the network safety level of the power network and reduce delay, and provides a more intelligent network resource allocation architecture for the future power interconnection network.
The technical scheme adopted by the invention is as follows: a secure and low-delay dual-layer network resource allocation method based on SDN/NFV network slicing technology comprises the following steps:
Step 1: taking a service strategy as an input, constructing a double-layer network resource allocation framework based on SDN/NFV network slicing technology;
Step 2: according to the double-layer network resource allocation architecture, deploying an SDN/NFV-based power grid infrastructure environment, defining end-to-end flow delay, service processing delay and data packet delivery time estimation, restricting a short path, and establishing an objective function for minimizing delay;
And 3, solving the objective function by adopting a K-means clustering algorithm to obtain an optimal intermediate box placement scheme, and providing a dynamic resource allocation method for predicting NIS (network identification) utilization rate based on regression tree analysis.
The first layer in the double-layer network resource allocation architecture is a power center cloud node and an SDN/NFV plane, and the second layer is a power network terminal.
The step 2 specifically comprises the following steps:
step 2.1: deploying an SDN/NFV-based power grid infrastructure environment:
SDN/NFV-based power grid infrastructure is represented as a directed graph Wherein/>Is a set of nodes, ε= { v i,vj } is a set of links, where i= {1,2,3,..N v } and j= {1,2,3,..N v } are subscripts of node pairs, and N v is the total number of nodes; SDN switch/>The maximum number of rules that can be stored is P s, so the number of rules currently stored in the switch flow table is denoted P s∈Ps; the set of all NIS's is noted asThen the NFV-based middlebox that provides these NIS is/>There may be N q middleboxes available in the network, which will/>Expressed as the number of NIS middleboxes, wherein/>Q is a rational number set; each intermediate box has a maximum processing capacity O b to execute a set of NIS;
The application data stream f can be described as Source n and dest n are the source node and destination node, respectively; /(I)Is the NIS that network traffic of a flow must access from source to destination; o n is the middlebox processing capacity occupied by NIS, t n is the daily clock period of the streaming request NIS; generating/>, after obtaining an application data stream For a group of streams that require NIS from the middlebox;
step 2.2: defining end-to-end flow delay, traffic processing delay and packet delivery time estimate, constraining short paths:
end-to-end flow delay is defined as:
wherein d if,bf is the aggregate delay time from the source ingress switch to the corresponding NIS middlebox, and d bf,ef is the aggregate delay time from the source ingress switch to the destination egress switch; the aggregate delay depends on the traffic processing delay Packet transfer time/>, per link between o n and two nodes per NIS n
The traffic processing delay and packet delivery time estimate are expressed as:
Where M is the number of application data streams requesting NIS; z max is the maximum data packet size, and the unit is bit; b r is the transmission bit rate, the unit is bit/s; Distance or length of the transmission medium is in meters; l s is the propagation speed in the medium in m/s;
The constrained short path is expressed as:
I.e. find one route R from the set of all routes R st, minimize the objective function f C (R) that minimizes the delay, such that the delay D (R) is less than or equal to the threshold D max; further, constraints can be varied, including traffic chain ratio, bandwidth consumption, deployment cost, energy consumption, and the like. However, whatever flow routing scheme is used, the impact of middlebox deployment on network latency is most pronounced. Therefore, there is a need to develop an appropriate deployment strategy to minimize each flow Is a function of the total delay of (a).
Step 2.3: this step will establish an objective function that minimizes the delay for subsequent optimization and solution.
The objective function to minimize the delay is expressed as:
the first constraint is to ensure that each intermediate box can be successfully connected to any SDN switch in the network; the second constraint is middleware processing capability to ensure that the corresponding middleware has the ability to handle NIS of application data flow requests; a third constraint is SDN switch memory capacity, which is used to confirm that the switch has available resources to store new rule entries.
The step 3 specifically comprises the following steps:
Step 3.1: solving the objective function by adopting a K-means clustering algorithm to obtain an optimal intermediate box placement scheme, which comprises the following specific steps:
First, collecting Shortest Path (SP) calculations between ingress switches per pair of flows; secondly, initializing the clusters using a careful seeding initialization process; wherein an SDN switch is used as a cluster center; then updating and verifying the center of each cluster to minimize the sum of SP delay times to reach all switches in the optimal number of clusters; checking and calculating the number of policies stored in each SDN switch each time; repeating the steps until the sub-network is divided into the optimal K sub-networks; finally, a NIS middle box is placed in each cluster center.
Step 3.2: providing a dynamic resource allocation method for predicting NIS utilization rate based on regression tree analysis;
After placing all NIS middleboxes in the optimal position, resources can be dynamically allocated to each NIS; in this case, the resource allocation of each service at a particular time depends on the proportion of services repeatedly requested by the application/user in the respective cluster, using the historical data of the application data stream as input, using a regression tree algorithm to predict the usage of the next time window, the NIS with higher predicted usage obtaining higher resource allocation at the next stage, and vice versa. Furthermore, to protect NIS from failures due to excessive and unpredictable requests, we have adopted the NFV-thread program.
Compared with the prior art, the invention has the beneficial effects that: the dual-layer network architecture based on SDN/NFV is constructed, high-level network security and low-delay communication are guaranteed, the optimization problem of a deployment middlebox is solved, NIS resource allocation is carried out based on the prediction of service utilization rate in each corresponding cluster, delay is minimized, and communication security and low delay are maintained.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention.
Detailed Description
In order to facilitate the understanding and practice of the invention, those of ordinary skill in the art will now make further details with reference to the drawings and examples, it being understood that the examples described herein are for the purpose of illustration and explanation only and are not intended to limit the invention thereto.
The present invention proposes an Artificial Intelligence (AI) driven solution consisting of two phases (nim middlebox placement based on optimized kmeans-based graph cluster analysis, dynamic resource allocation based on regression tree analysis to predict NIS usage) that helps to provide a secure and low-latency SDN/NFV based network infrastructure for the grid. The method for distributing the safe and low-delay double-layer network resources based on SDN/NFV network slicing technology provided by the invention, referring to figure 1, comprises the following steps:
Step 1: constructing a double-layer network resource allocation architecture based on SDN/NFV network slicing technology;
Step 2: according to the double-layer network resource allocation architecture, deploying an SDN/NFV-based power grid infrastructure environment, defining end-to-end flow delay, service processing delay and data packet delivery time estimation, restricting a short path, and establishing an objective function for minimizing delay;
Step 3, providing an NIS middle box placement method based on optimized kmeans-based graph cluster analysis and dynamic resource allocation of predicted NIS utilization rate based on regression tree analysis by using the objective function in the step 2;
The step 1 comprises the following steps: under the network security background, taking a service strategy as an input, virtually applying a set of artificial intelligent driven Network Intelligent Services (NIS) to a Network Function Virtualization (NFV) -based middlebox covered on a Software Defined Network (SDN) architecture, and constructing a SDN/NFV-based double-layer network resource allocation architecture, wherein a first layer in the double-layer network resource allocation architecture is an electric power center cloud node and an SDN/NFV plane; the second layer is a power network terminal.
The step 2 comprises the following steps:
Step 2.1: the step is to deploy an SDN/NFV-based power grid infrastructure environment;
SDN/NFV-based grid infrastructure is represented as a simple directed graph Wherein the method comprises the steps ofIs a set of nodes, ε= { v i,vj } is a set of links, where i= {1,2,3,..N v } and j= {1,2,3,..N v } are subscripts of node pairs and N v is the total number of nodes. SDN switch/>The maximum number of rules that can be stored is P s, so the number of rules currently stored in the switch flow table is denoted P s∈Ps. If the set of all NIS is recorded as/>Then the NFV-based middlebox that provides these services isThere may be N q middleboxes available in the network, so we will/>Expressed as the number of NIS middleboxes, whereQ is a rational number set. Each middlebox has a maximum processing capacity O b for executing a set of NIS. This processing capability is expressed in units of Mbps depending on the Central Processing Unit (CPU) available in each middlebox.
The application data stream f can be described asSource n and dest n are the source node and destination node, respectively. /(I)Is the NIS that network traffic of a flow must access from source to destination. Where o n is the middlebox processing capacity occupied by the NIS and t n is the daily clock period of the streaming request NIS. After knowing the application data stream, we can generate/>This is a set of streams that require NIS from the middlebox.
Step 2.2: this step will define end-to-end flow delays, traffic processing delays and packet delivery time estimates, constraining short paths.
First define the end-to-end flow delay as
Where d if,bf is the aggregate delay time from the source ingress switch to the corresponding NIS middlebox and d bf,ef is the aggregate delay time from the source ingress switch to the destination egress switch. The aggregate delay depends on the traffic processing delayThe time/>, of packet delivery per link between o n of each NIS n (i.e., the nth NIS) and two nodes
The traffic processing delay and packet delivery time estimate are expressed as
Where M is the number of application data streams requesting NIS. For delay-sensitive applications, satisfaction follows an s-type utility function. Thus, an accurate resource allocation policy is an unavoidable task to improve QoS. Wherein Z max is the maximum data packet size, and the unit is bit; b r is the transmission bit rate, the unit is bit/s; Distance or length of the transmission medium is in meters; l s is the propagation speed in the medium in m/s. Propagation speed depends on the physical medium of the link, e.g. copper wire is 2 x 108 m/s and wireless communication is 3 x 108 m/s.
Constrained short paths are expressed as
I.e. one route R is found from the set of all routes R st, the objective function f C (R) is minimized such that the delay D (R) is less than or equal to the threshold D max. Further, constraints can be varied, including traffic chain ratio, bandwidth consumption, deployment cost, energy consumption, and the like. However, whatever flow routing scheme is used, the impact of middlebox deployment on network latency is most pronounced. Therefore, there is a need to develop an appropriate deployment strategy to minimize each flowIs a function of the total delay of (a).
Step 2.3: this step will establish an objective function that minimizes the delay for subsequent optimization and solution.
Delay minimization problem is expressed as
There are three constraints on this problem, the first constraint being to ensure that each middlebox can successfully connect to any one SDN switch in the network. The second constraint is middleware processing capability to ensure that the corresponding middleware has the ability to handle NIS of application data flow requests. A third constraint is SDN switch memory capacity, which is used to confirm that the switch has available resources to store new rule entries.
And 3, providing a graph clustering analysis method for middle box placement.
Under the condition of avoiding using a threshold method for graph clustering analysis, adopting a K-means clustering algorithm, and considering the following additional conditions:
1) Since the goal is to find the NIS middlebox placement with minimal latency, the cluster center initialization method plays a key role. Thus, an initialization and careful seed selection process is essential.
2) The recalculated cluster center should be selected from the SDN ingress switches. Then, a cluster refinement process is performed, reassigning all SDN switches to the appropriate clusters.
3) The distance calculation method also needs to take into account nodes that are indirectly connected or that are not physically connected to the cluster center.
The step 3 specifically comprises the following steps:
Step 3.1: solving the objective function by adopting a K-means clustering algorithm to obtain an optimal intermediate box placement scheme, which comprises the following specific steps:
First, collecting Shortest Path (SP) calculations between ingress switches per pair of flows; secondly, initializing the clusters using a careful seeding initialization process; furthermore, much like the K-means clustering approach, we propose a graph clustering based middlebox placement approach using selected nodes, where SDN switches are used as the cluster centers; rather than using the nearest average. Then updating and verifying the center of each cluster to minimize the sum of SP delay times to reach all switches in the optimal number of clusters; taking additional conditions into account, the number of policies stored in each SDN switch needs to be checked and calculated each time; repeating the steps until the sub-network is divided into the optimal K sub-networks; finally, a NIS middle box is placed in each cluster center.
Step 3.2: this step will propose a corresponding dynamic resource allocation method.
After placing all NIS middleboxes in the optimal position, resources can be dynamically allocated to each NIS; in this case, the resource allocation of each service at a particular time depends on the proportion of services repeatedly requested by the application/user in the respective cluster, using the historical data of the application data stream as input, using a regression tree algorithm to predict the usage of the next time window, the NIS with higher predicted usage obtaining higher resource allocation at the next stage, and vice versa. Furthermore, to protect NIS from failures due to excessive and unpredictable requests, we have adopted the NFV-thread program.
It should be understood that parts of the specification not specifically set forth herein are all prior art.
It should be understood that the foregoing description of the preferred embodiments is not intended to limit the scope of the invention, but rather to limit the scope of the claims, and that those skilled in the art can make substitutions or modifications without departing from the scope of the invention as set forth in the appended claims.
Claims (4)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311872057.6A CN117914773A (en) | 2023-12-30 | 2023-12-30 | A secure and low-latency two-layer network resource allocation method based on SDN/NFV network slicing technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311872057.6A CN117914773A (en) | 2023-12-30 | 2023-12-30 | A secure and low-latency two-layer network resource allocation method based on SDN/NFV network slicing technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117914773A true CN117914773A (en) | 2024-04-19 |
Family
ID=90691668
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311872057.6A Pending CN117914773A (en) | 2023-12-30 | 2023-12-30 | A secure and low-latency two-layer network resource allocation method based on SDN/NFV network slicing technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117914773A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120358190A (en) * | 2025-06-25 | 2025-07-22 | 武汉大学 | Optimization method and device for multi-mode network global route, electronic equipment and medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020181403A1 (en) * | 2001-05-31 | 2002-12-05 | Nec Corporation | Communication path designing method, communication path designing device, and program to have computer execute same method |
| CN108206790A (en) * | 2018-01-11 | 2018-06-26 | 重庆邮电大学 | A kind of selection of SDN joint routes and resource allocation methods based on network slice |
| US20200099625A1 (en) * | 2018-09-24 | 2020-03-26 | Netsia, Inc. | Path determination method and system for delay-optimized service function chaining |
| CN110971451A (en) * | 2019-11-13 | 2020-04-07 | 国网河北省电力有限公司雄安新区供电公司 | NFV resource allocation method |
| CN111865681A (en) * | 2020-07-14 | 2020-10-30 | 中国电力科学研究院有限公司 | End-to-end delay optimization method, system and storage medium for core network slicing |
| CN111865668A (en) * | 2020-06-30 | 2020-10-30 | 南京邮电大学 | A network slicing method based on SDN and NFV |
| CN112491619A (en) * | 2020-11-25 | 2021-03-12 | 东北大学 | Self-adaptive distribution technology for service customized network resources based on SDN |
| CN114205316A (en) * | 2021-12-31 | 2022-03-18 | 全球能源互联网研究院有限公司 | Network slice resource allocation method and device based on power service |
-
2023
- 2023-12-30 CN CN202311872057.6A patent/CN117914773A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020181403A1 (en) * | 2001-05-31 | 2002-12-05 | Nec Corporation | Communication path designing method, communication path designing device, and program to have computer execute same method |
| CN108206790A (en) * | 2018-01-11 | 2018-06-26 | 重庆邮电大学 | A kind of selection of SDN joint routes and resource allocation methods based on network slice |
| US20200099625A1 (en) * | 2018-09-24 | 2020-03-26 | Netsia, Inc. | Path determination method and system for delay-optimized service function chaining |
| CN110971451A (en) * | 2019-11-13 | 2020-04-07 | 国网河北省电力有限公司雄安新区供电公司 | NFV resource allocation method |
| CN111865668A (en) * | 2020-06-30 | 2020-10-30 | 南京邮电大学 | A network slicing method based on SDN and NFV |
| CN111865681A (en) * | 2020-07-14 | 2020-10-30 | 中国电力科学研究院有限公司 | End-to-end delay optimization method, system and storage medium for core network slicing |
| CN112491619A (en) * | 2020-11-25 | 2021-03-12 | 东北大学 | Self-adaptive distribution technology for service customized network resources based on SDN |
| CN114205316A (en) * | 2021-12-31 | 2022-03-18 | 全球能源互联网研究院有限公司 | Network slice resource allocation method and device based on power service |
Non-Patent Citations (2)
| Title |
|---|
| ARDIANSYAH ET AL: "Latency-Optimal Network Intelligence Services in SDN/NFV-Based Energy Internet Cyberinfrastructure", SPECIAL SECTION ON SOFTWARE DEFINED NETWORKS FOR ENERGY INTERNET AND SMART GRID COMMUNICATIONS, 24 October 2019 (2019-10-24) * |
| 徐冉;王文东;龚向阳;阙喜戎;: "网络功能虚拟化中延时感知的资源调度优化方法", 计算机研究与发展, no. 04, 15 April 2018 (2018-04-15) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN120358190A (en) * | 2025-06-25 | 2025-07-22 | 武汉大学 | Optimization method and device for multi-mode network global route, electronic equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2023039965A1 (en) | Cloud-edge computing network computational resource balancing and scheduling method for traffic grooming, and system | |
| US11516146B2 (en) | Method and system to allocate bandwidth based on task deadline in cloud computing networks | |
| CN105960783B (en) | Inter-domain SDN traffic engineering | |
| CN101707788B (en) | Differential pricing strategy based dynamic programming method of multilayer network services | |
| CN108566659A (en) | A kind of online mapping method of 5G networks slice based on reliability | |
| Buyakar et al. | Resource allocation with admission control for GBR and delay QoS in 5G network slices | |
| EP2225851B1 (en) | Improved resource allocation plan in a network | |
| Mahapatra et al. | Utilization-aware VB migration strategy for inter-BBU load balancing in 5G cloud radio access networks | |
| Nouruzi et al. | Online service provisioning in NFV-enabled networks using deep reinforcement learning | |
| CN115002799A (en) | A task offloading and resource allocation method for industrial hybrid networks | |
| CN114710196A (en) | Software-defined satellite network virtual network function migration method | |
| CN108092895A (en) | A kind of software defined network joint route selection and network function dispositions method | |
| CN117914773A (en) | A secure and low-latency two-layer network resource allocation method based on SDN/NFV network slicing technology | |
| Tzanakaki et al. | A converged network architecture for energy efficient mobile cloud computing | |
| Zhao et al. | On the parallel reconfiguration of virtual networks in hybrid optical/electrical datacenter networks | |
| Wu et al. | Resource allocation optimization in the NFV-enabled MEC network based on game theory | |
| Zhu et al. | Efficient hybrid multicast approach in wireless data center network | |
| CN108880895B (en) | SDN joint routing selection and network function deployment method based on user stream transmission cost optimization | |
| Luo et al. | Traffic-aware VDC embedding in data center: A case study of fattree | |
| Lv et al. | Mobile edge computing oriented multi-agent cooperative routing algorithm: A DRL-based approach | |
| CN117858095A (en) | Wireless access network slicing method based on group learning of DTA in Internet of things | |
| CN115277531B (en) | A Two-Phase Routing Method for Multipath Bottleneck Fairness Constraints in Cloud Wide Area Networks | |
| CN114221948B (en) | Cloud network system and task processing method | |
| CN106850726B (en) | Load-aware request routing method for cloud data center based on SDN | |
| Guanqiang et al. | A method for IOT device management and traffic scheduling in Distribution Station area based on Distributed Sdn Architecture |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Xu Huan Inventor after: Zhou Dekun Inventor after: Meng Haohua Inventor after: Zheng Lei Inventor after: Long Fei Inventor after: Xia Fan Inventor after: Zhao Jingyao Inventor after: Wei Xiaoyan Inventor after: Mei Ziwei Inventor after: Wang Hongwei Inventor after: Zeng Zheng Inventor after: Zhou Zheng Inventor after: Wang Yixi Inventor after: Li Lei Inventor after: Wang Chengwei Inventor after: Chen Qimei Inventor after: Jin Bo Inventor after: Chen Jialin Inventor after: Gao Fei Inventor after: Wang Weiran Inventor after: Zha Zhiyong Inventor after: Yu Zheng Inventor before: Chen Jialin Inventor before: Meng Haohua Inventor before: Zheng Lei Inventor before: Long Fei Inventor before: Xu Huan Inventor before: Xia Fan Inventor before: Zhao Jingyao Inventor before: Wei Xiaoyan Inventor before: Mei Ziwei Inventor before: Wang Hongwei Inventor before: Zeng Zheng Inventor before: Zhou Zheng Inventor before: Wang Yixi Inventor before: Li Lei Inventor before: Wang Chengwei Inventor before: Chen Qimei Inventor before: Jin Bo Inventor before: Zhou Dekun Inventor before: Zha Zhiyong Inventor before: Wang Weiran Inventor before: Yu Zheng Inventor before: Gao Fei |
|
| CB03 | Change of inventor or designer information |