WO2022252496A1 - 软件定义云边协同网络能耗优化方法和系统 - Google Patents

软件定义云边协同网络能耗优化方法和系统 Download PDF

Info

Publication number
WO2022252496A1
WO2022252496A1 PCT/CN2021/129564 CN2021129564W WO2022252496A1 WO 2022252496 A1 WO2022252496 A1 WO 2022252496A1 CN 2021129564 W CN2021129564 W CN 2021129564W WO 2022252496 A1 WO2022252496 A1 WO 2022252496A1
Authority
WO
WIPO (PCT)
Prior art keywords
energy consumption
computing
cloud
node
edge
Prior art date
Application number
PCT/CN2021/129564
Other languages
English (en)
French (fr)
Inventor
陈伯文
黄守翠
梁瑞鑫
刘玲
吴金炳
刘秀敏
Original Assignee
苏州路之遥科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 苏州路之遥科技股份有限公司 filed Critical 苏州路之遥科技股份有限公司
Publication of WO2022252496A1 publication Critical patent/WO2022252496A1/zh

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
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/25—Arrangements specific to fibre transmission
    • 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
    • H04L41/06—Management of faults, events, alarms or notifications
    • H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
    • 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
    • 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
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/04—Wireless resource allocation
    • H04W72/044—Wireless resource allocation based on the type of the allocated resource
    • H04W72/0453—Resources in frequency domain, e.g. a carrier in FDMA
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Definitions

  • the invention relates to the technical field of cloud-edge collaborative network, in particular to a method and system for optimizing energy consumption of a software-defined cloud-edge collaborative network.
  • a new architecture and technology mobile cloud computing has the potential to address the above challenges.
  • cloud computing can provide a large number of mobile application access functions, and reduce network energy consumption through the unified management of cloud computing.
  • Devices can be offloaded by computing resources, and the Computing tasks are transmitted to remote cloud servers for execution, which can effectively alleviate the problem of large demand for computing resources.
  • the transmission of computing tasks to cloud servers will cause unacceptable delays and increase energy consumption for additional transmissions.
  • the traditional network cannot control the traffic from the perspective of the entire network, the flexible adjustment capability of the traffic path is insufficient, and the network operator is responsible for providing the network access function for the user. It needs to be upgraded to the level of protocol formulation and modification, which is difficult to operate and maintain, and lacks the scalability of the network.
  • an embodiment of the present invention provides a method and system for optimizing energy consumption of a software-defined cloud-edge collaborative network to solve the problem of insufficient flexibility and scalability in the cloud-edge collaborative network in the prior art.
  • An embodiment of the present invention provides a method for optimizing energy consumption of a software-defined cloud-edge collaborative network, including:
  • the optimal working path is obtained through the K shortest path algorithm.
  • obtaining a set of connection requests includes:
  • connection request sets CR Generate a set of connection request sets CR; among them, each connection request CR(u,f,r) ⁇ CR, u represents the user request, f represents the number of spectrum slots required by the user request, and r represents the computing resources required by the user request number.
  • determining the computing resource according to the connection request set includes:
  • the computing resource in the local area processes the set of connection requests
  • the local area computing resource and the edge area computing resource process the connection request set
  • the computing resource in the local area and the computing resource in the edge area do not satisfy the connection request set, the computing resource in the local area, the computing resource in the edge area and the cloud computing resource process the connection request set.
  • obtaining the optimal working path through the K shortest path algorithm based on service transmission energy consumption, node energy consumption and computing energy consumption includes:
  • calculating the number of load tasks, node load and node load rate of each node includes:
  • calculating the sending energy consumption and node energy consumption of the K working paths includes:
  • P u represents the transmission power of the user
  • S u represents the size of the task
  • r represents the fiber transmission rate
  • J represents the number of nodes passed by the path
  • y j represents the load rate of the j-th node device port.
  • obtaining computing energy consumption according to the number of layers requested by the user includes:
  • the computing energy consumption is:
  • the computing energy consumption is:
  • ⁇ f represents the real-time processing capability of the edge computing service layer; Indicates the energy consumption generated by the edge computing service layer processing data per unit time; ⁇ c indicates the processing power allocated to each task by the cloud computing service layer; Indicates the energy consumption per unit time of processing data at the cloud computing service layer.
  • the spectrum allocation algorithm of the first hit is adopted, and the spectrum resource table is generated and numbered according to the spectrum resource status of all links on the candidate path;
  • the transmission energy consumption is calculated according to the user's sending power, task size and fiber transmission rate
  • the path node energy consumption is calculated according to the number of nodes passed by the path and the load of the nodes, and according to the calculation of the computing edge server Capacity and energy consumption per unit time of data processing by edge computing servers, and calculate the energy consumption of the software-defined cloud-edge collaborative network.
  • the software-defined central controller issues a command set, and uses the K shortest path algorithm to calculate the service arrival through the dynamically changing topology information of the cloud-edge collaborative network.
  • Work paths between edge computing servers After the working path is successfully selected, according to the resource occupancy status information mastered by the software-defined central controller, using the OpenFlow extension protocol, the software-defined central controller issues a resource allocation command, and uses the first-hit spectrum allocation algorithm to perform spectrum allocation on the path.
  • Resource allocation needs to meet the two constraints of spectrum consistency and spectrum continuity at the same time, and then update the status of network computing resources and spectrum resources in real time, which improves the flexibility of the system.
  • the embodiment of the present invention also provides a software-defined cloud-edge collaborative network energy consumption optimization system, including:
  • the centralized control module is used to dynamically understand and adjust the resource information of each node in real time.
  • the centralized control module adopts the centralized control and management mode based on the OpenFlow extension protocol to manage the cloud-edge collaborative network and establish a software-defined cloud-edge collaborative network;
  • the application service requirement module is used to receive the command set issued by the centralized control and management module of the centralized control module according to the application service requirements and business service level requirements in the software-defined cloud-edge collaborative network, establish the working path and allocate the required spectrum resources, complete data transmission required by application services, update computing resources and release network resources occupied by application requirements.
  • transmit energy consumption is calculated according to the user’s sending power, task size and fiber transmission rate
  • energy consumption of path nodes is calculated according to the number of nodes passed by the path and the load of the nodes, and according to the computing power and
  • the edge computing server processes the energy consumption per unit time of data, and calculates the energy consumption of the software-defined cloud-edge collaboration network.
  • Fig. 1 shows a flow chart of a software-defined cloud-edge collaborative network energy consumption optimization method in an embodiment of the present invention
  • FIG. 2 shows a flowchart of a method for optimizing energy consumption of a software-defined cloud-edge collaborative network in an embodiment of the present invention
  • Fig. 3 shows a network model diagram of a software-defined cloud-edge collaborative network energy consumption optimization system in an embodiment of the present invention
  • Fig. 4 shows a structural diagram of a software-defined cloud-edge collaborative network energy consumption optimization system in an embodiment of the present invention.
  • an embodiment of the present invention provides a method for optimizing energy consumption of a software-defined cloud-edge collaborative network, including:
  • Step S10 acquiring a set of connection requests.
  • This step includes: generating a set of connection request sets CR; wherein, each connection request CR(u, f, r) ⁇ CR, u represents the user request, f represents the number of spectrum gaps required by the user request, and r represents the number of spectrum gaps required by the user request The number of computing resources required.
  • Step S20 determining computing resources according to the connection request set.
  • the software-defined central controller determines whether the local area where the user request is located has sufficient computing resources.
  • the software-defined central controller issues a command to send the user request to the local computing server for processing.
  • the user request does not pass through other nodes in the network, and the system consumes only It is enough to consider the energy consumption of sending to the edge server in the local area and the calculation energy consumption of the edge server in the local area. If the computing resources in the local area are insufficient, the software-defined central controller makes a decision to migrate the computing to the edge computing server or cloud computing server.
  • Step S30 based on the service transmission energy consumption, node energy consumption and calculation energy consumption, the optimal working path is obtained through the K shortest path algorithm.
  • the K shortest path algorithm is used to calculate the working path from the user request to the edge computing server.
  • the K shortest path algorithm calculates K candidate paths as routing options, and sorts the energy consumption of each path from small to large, that is, the smaller the energy consumption, the higher the priority. If the path ranked first is blocked on a certain link, the paths ranked later are sequentially selected for spectrum resource allocation until resources are allocated successfully, or all paths are blocked.
  • transmit energy consumption is calculated according to the user’s sending power, task size and fiber transmission rate
  • energy consumption of path nodes is calculated according to the number of nodes passed by the path and the load of the nodes, and according to the computing power and
  • the edge computing server processes the energy consumption per unit time of data, and calculates the energy consumption of the software-defined cloud-edge collaboration network.
  • determining the computing resource according to the connection request set includes:
  • the computing resource in the local area processes the set of connection requests
  • the local area computing resource and the edge area computing resource process the connection request set
  • the computing resource in the local area and the computing resource in the edge area do not satisfy the connection request set, the computing resource in the local area, the computing resource in the edge area and the cloud computing resource process the connection request set.
  • the software-defined central server determines whether the server in the edge area has enough computing resources to process user requests. If the computing resources of the server in the edge region meet the computing resources required by the user request, the software-defined central controller issues a command to the adjacent server, and needs to migrate the user request to other regional edge computing servers on the same switch in the local region.
  • the system energy consumption includes the energy consumption of sending user requests to adjacent servers, the energy consumption of nodes generated when passing through switches, and the computing energy consumption of adjacent edge servers.
  • the software-defined central controller detects that the computing resources on the local and adjacent edge computing area servers cannot meet the computing resources required by the user's request, it will then determine whether the cloud server has sufficient computing resources. If the cloud server has sufficient computing resources, the user request is sent to the cloud server through the switch for data processing. At this time, the system energy consumption includes the transmission energy consumption of the user sending the request to the cloud server, the node energy consumption and the computing energy consumption of the cloud server.
  • obtaining the optimal working path through the K shortest path algorithm based on the service transmission energy consumption, node energy consumption and calculation energy consumption includes:
  • the number of load tasks X j (t) of any node j is calculated:
  • b, i, and s are all natural numbers greater than or equal to 1.
  • P u represents the transmission power of the user
  • S u represents the size of the task
  • r represents the fiber transmission rate
  • J represents the number of nodes passed by the path
  • y j represents the load rate of the j-th node device port.
  • the computing energy consumption is:
  • the computing energy consumption is:
  • ⁇ f represents the real-time processing capability of the edge computing service layer; Indicates the energy consumption generated by the edge computing service layer processing data per unit time; ⁇ c indicates the processing power allocated to each task by the cloud computing service layer; Indicates the energy consumption per unit time of processing data at the cloud computing service layer.
  • the spectrum allocation algorithm of the first hit is adopted, and the spectrum resource table is generated and numbered according to the spectrum resource status of all links on the candidate path;
  • connection request CR(u,f,r) After the connection request CR(u,f,r) successfully establishes the working path, it notifies the software-defined central controller that the data service requested by the user can be transmitted, and the spectrum resources of the working path are allocated according to the constraints of spectrum consistency and spectrum continuity. distribute.
  • a first-hit spectrum allocation algorithm is adopted, and a spectrum resource table is generated and numbered according to the spectrum resource states of all links on the path, and available spectrum gaps are searched from the end with the smaller number. If an available spectrum gap is found, spectrum resource allocation and spectrum status update are performed; if no spectrum gap is found, spectrum allocation fails and services are blocked.
  • the optimal working path and allocating spectrum resources after obtaining the optimal working path and allocating spectrum resources, it further includes: updating the computing resources in the edge area, and recording the number of successful connection establishment requests.
  • the software-defined central controller issues a command set, and uses the K shortest path algorithm to calculate the service to the edge computing through the dynamically changing topology information of the cloud-edge collaborative network Work paths between servers.
  • the software-defined central controller issues a resource allocation command, and uses the first-hit spectrum allocation algorithm to perform spectrum processing on the path.
  • Resource allocation needs to meet the two constraints of spectrum consistency and spectrum continuity at the same time, and then update the status of network computing resources and spectrum resources in real time, which improves the flexibility of the system.
  • An embodiment of the present invention provides a software-defined cloud-edge collaborative network energy consumption optimization system, including a centralized control module and an application service demand module, wherein: the centralized control module is used to dynamically understand and adjust the resource information of each node in real time, and centrally control The module adopts the centralized control and management mode based on the OpenFlow extension protocol to manage the cloud-edge collaborative network, and establishes a software-defined cloud-edge collaborative network; the application service requirement module is used in the software-defined cloud-edge collaborative network, based on application service requirements and business service level requirements , receiving the command set issued by the centralized control and management module of the centralized control module, establishing the working path and allocating the required spectrum resources, completing the data transmission required by the application service, updating the computing resources and releasing the network resources occupied by the application requirements.
  • the centralized control module mainly includes:
  • This module mainly obtains the resource information and network topology information of all cloud-edge collaborative networks, monitors the status of each node, and can calculate the transmission path according to business needs, and allocate and release corresponding services for application services. network resources.
  • Cloud-edge collaborative network initialization module initialize the cloud-edge collaborative network through the OpenFlow extension protocol, and configure network topology information, optical network connection status, and user requests in the cloud-edge collaborative network G(U,B,I,S) number, number of edge computing servers, number of base stations and switches.
  • Network programmable hardware module the main function of this module is to introduce a programmable hardware module, which is applied to the monitoring of cloud-edge collaborative network control status and perception of network resource usage status.
  • Embedded software development module which can embed the developed software program and dynamically modify it according to the needs of the cloud-edge collaborative network. At the same time, it can also change the status according to business needs and modify the status parameters of the software and other information.
  • Network status monitoring module mainly completes the status of spectrum flexible optical network initialization, connection request generation, service priority selection, edge computing server selection, working path establishment, spectrum resource allocation, computing resource update, resource release, and network energy consumption calculation Monitoring capabilities to achieve the goal of minimizing system energy consumption when computing resource allocation.
  • Judgment and early warning module Execute the coordination function between various modules, and whether each module has established a successful judgment and early warning function, and complete the goal of reducing system energy consumption in mobile edge computing.
  • the application service requirements module mainly includes:
  • This module is mainly for service requests sent by users according to requirements, and information such as the number of configuration connection requests, the number of spectrum gaps required by different connection requests, and computing resources.
  • the command set module of the software-defined central controller this module can complete the distribution of application requirements according to the resource information status of the cloud-edge collaborative network and the command set executed in the centralized control module based on the OpenFlow extension protocol in the centralized control module mechanism.
  • —Working path establishment module According to the user request of the connection request CR(u,f,r) and the edge computing server processing the request, the K shortest path algorithm is used to calculate K candidate paths from the user request to the server, so as to find The optimal path is selected as the working path.
  • Specific resource allocation module according to the number of spectrum gaps f required by the connection request CR(u,f,r), search for the bandwidth resources required to meet the connection request in the selected working path, if both spectrum continuity and spectrum If the dual constraint conditions of consistency are met, the connection request is established successfully; if the dual constraint conditions of spectrum continuity and spectrum consistency cannot be satisfied at the same time, the connection request establishment fails.
  • Compute resource update module After the spectrum resources are successfully allocated, the edge computing server processing user requests is updated in real time with computing resources.
  • Resource release module after the connection request is successfully transmitted, the resource release is performed on the spectrum resource occupied by the working path. At the same time, the computing resources of the edge computing server processing the user request are released. Finally, the information of the working path established by the connection request is cleared.
  • —Energy consumption calculation module After each connection request is successfully established, record the node information passed by each connection request transmission path, and calculate the user sending energy consumption and node energy consumption of each connection request according to the formula in embodiment 1 , and then record the location of the edge computing server determined by the user request, and calculate the computing energy consumption of each connection request.
  • layer 1 has three local areas, and each area has base stations and corresponding edge computing servers.
  • Layer 2 is a cloud network composed of switches and corresponding edge computing servers.
  • the programming network nodes of the OpenFlow agent are integrated, such as the programmable nodes 1, 2, 3, 4, 5, 6, 7, 8, 9;
  • each node is controlled by a software-defined central controller, and a software-defined cloud-edge collaborative network is established.
  • the programmable nodes A and C discover the response command and inform the central controller, as shown in Figure 34, the path requested by the user is successfully established, and the data service transmission required by the service can be carried out. After the data transmission is completed, the resource is released and the connection Request to clear the path established;
  • the core In order to establish a software-defined cloud-edge collaborative network, carry the data transmission requested by users, and ensure the safe and efficient operation of user requests in the software-defined cloud-edge collaborative network, the core is to introduce programmable functions into network nodes, through a centralized controller based on the OpenFlow extension protocol , which can monitor the real-time status of the entire software-defined cloud-edge collaboration network. According to the attribute requirements of application services, the transmission path of the cloud-edge collaborative network is established to complete the end-to-end transmission of business requirements.
  • the protocol performs centralized control and management, simplifies the resource scheduling and optimization of the cloud-edge collaborative network, and realizes a software-defined cloud-edge collaborative network energy optimization method.
  • the storage medium can be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a flash memory (Flash Memory), a hard disk (Hard Disk) Disk Drive, abbreviation: HDD) or solid-state hard disk (Solid-State Drive, SSD) etc.;
  • the storage medium can also include the combination of the above-mentioned types of memory.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明公开了一种软件定义云边协同网络能耗优化方法及系统，其中方法包括：获取连接请求集合；根据连接请求集合确定计算资源；基于业务发送能耗、节点能耗和计算能耗通过K条最短路径算法获取最优工作路径。根据用户发送功率、任务大小和光纤传输速率计算发射能耗，根据路径所经节点数和节点的负载情况计算路径节点能耗，以及根据计算边缘服务器的计算能力和边缘计算服务器处理数据单位时间所产生的能耗，计算出软件定义云边协同网络的能耗。通过软件定义云边协同网络能耗优化方法和系统，有效平衡网络资源和业务能耗之间的关系，在合理分配资源的情况下最大程度地减少云边协同网络的能耗，提高了云边协同网络的灵活性和可拓展性。

Description

软件定义云边协同网络能耗优化方法和系统 技术领域
本发明涉及云边协同网络技术领域，具体涉及一种软件定义云边协同网络能耗优化方法和系统。
背景技术
近年来，随着物联网(Internet of things，IoT)的高速发展和大量数据应用的广泛使用，用户对于网络计算资源的需求量急剧增长。资源受限的移动终端设备的计算能力已不能满足移动用户在数据处理方面急剧增加的需求。尽管新的移动设备在中央处理器(Central Processing Unit，CPU)方面越来越强大，但移动应用程序，如交互式游戏、虚拟现实和自然语言处理，通常需要进行密集的计算，并产生高能耗。
一个新的体系结构和技术移动云计算具有应对上述挑战的潜力。通过将计算任务从移动设备迁移到基于基础架构的云服务器，云计算可以提供大量的移动应用程序接入功能，通过云计算的统一管理方式降低网络的能耗，设备可以通过计算资源卸载，将计算任务传输到远端云服务器执行，从而能够有效缓解计算资源需求量较大的问题。然而，计算任务传输到云端服务器会造成不可接受的时延、增加额外传输能量消耗问题。由于传统网络不能从整个网络的角度对流量进行控制，流量路径的灵活调整能力不足，网络运营商负责对用户提供网络接入功能，用户的需求千差万别，一旦原有基础网络无法满足新需求，就需要上升到协议制定修改的层面，运维难度较大，缺乏网络的可扩展性。
因此，如何提高云边协同网络的灵活度和可扩展性成为亟待解决的问题。
发明内容
有鉴于此，本发明实施例提供了一种软件定义云边协同网络能耗优化方法和系统，以解决现有技术中云边协同网络存在的灵活度和可扩展性不够高的问题。
本发明实施例提供了一种软件定义云边协同网络能耗优化方法，包括：
获取连接请求集合；
根据连接请求集合确定计算资源；
基于业务发送能耗、节点能耗和计算能耗通过K条最短路径算法获取最优工作路径。
可选地，获取连接请求集合包括：
生成一组连接请求集合CR；其中，每一个连接请求CR(u,f,r)∈CR，u表示用户请求，f表示用户请求所需的频谱间隙数，r表示用户请求所需的计算资源数。
可选地，根据连接请求集合确定计算资源包括：
若本地区域计算资源满足连接请求集合，则由本地区域计算资源处理连接请求集合；
若本地区域计算资源不满足连接请求集合，则由本地区域计算资源和边缘区域计算资源处理连接请求集合；
若本地区域计算资源和边缘区域计算资源不满足连接请求集合，则由本地区域计算资源、边缘区域计算资源和云计算资源处理连接请求集合。
可选地，基于业务发送能耗、节点能耗和计算能耗通过K条最短路径算法获取最优工作路径包括：
计算每一节点的负载任务数、节点负载和节点负载率；
计算K条工作路径的发送能耗和节点能耗；
根据用户请求所在层数获取计算能耗；
按照最低能耗对K条工作路径进行排序，得到多个备选路径；
按照能耗由低到高检测、排除存在阻塞的备选路径，得到最优工作路径，并进行频谱资源分配。
可选地，计算每一节点的负载任务数、节点负载和节点负载率包括：
计算任一节点j的负载任务数X j(t)：
Figure PCTCN2021129564-appb-000001
计算任一节点j的负载Xc j(t)：
Figure PCTCN2021129564-appb-000002
计算任一节点j的负载率y j：
Figure PCTCN2021129564-appb-000003
其中，x l,j(t)表示时隙t由节点l到达节点j的任务数；
Figure PCTCN2021129564-appb-000004
表示节点j中第k个任务的大小；TR j表示节点j的资源总和；B＝{1,2,…,b,…}表示一组基站；I＝{1,2,…,i,…}表示一组交换机；S＝B+I＝{1,2,…,s,…}表示一组边缘计算服务器可选取的位置。
可选地，计算K条工作路径的发送能耗和节点能耗包括：
计算发送能耗SEE：
Figure PCTCN2021129564-appb-000005
计算节点能耗NOE：
Figure PCTCN2021129564-appb-000006
其中，P u表示用户的发送功率；S u表示任务的大小；r表示光纤传输速率；J表示路径经过的节点数；
Figure PCTCN2021129564-appb-000007
表示第j个节点设备端口空载功耗；
Figure PCTCN2021129564-appb-000008
表示第j个节点设备端口满载功耗；y j表示第j个节点设备端口负载率。
可选地，根据用户请求所在层数获取计算能耗包括：
若用户请求在边缘计算服务层，则计算能耗为：
Figure PCTCN2021129564-appb-000009
若用户请求在云计算服务层，则计算能耗为：
Figure PCTCN2021129564-appb-000010
其中，δ f表示边缘计算服务层的实时处理能力；
Figure PCTCN2021129564-appb-000011
表示边缘计算服务层处理数据单位时间所产生的能耗；δ c表示云计算服务层分配给每个任务的处理能力；
Figure PCTCN2021129564-appb-000012
表示云计算服务层处理数据单位时间所产生的能耗。
可选地，在得到多个备选路径之后，还包括：
基于频谱一致性和频谱连续性的约束条件，采用首次命中的频谱分配算法，根据备选路径上所有链路的频谱资源状态生成频谱资源表并进行编号；
从标号小的一端开始查找可用的频谱间隙；若当前频谱间隙可用，则进行频谱资源分配，并更新频谱状态；若未找到可用频谱，则频谱资源分配失败，业务存在阻塞。
可选地，在得到最优工作路径，并进行频谱资源分配之后，还包括：
对边缘区域计算资源进行更新，记录成功建立连接请求数。
本发明实施例的有益效果：
1.每一个连接请求成功建立后，根据用户发送功率、任务大小和光纤传输速率计算发射能耗，根据路径所经节点数和节点的负载情况计算路径节点能耗，以及根据计算边缘服务器的计算能力和边缘计算服务器处理数据单位时间所产生的能耗，计算出软件定义云边协同网络的能耗。通过软件定义云边协同网络能耗优化方法和系统，有效平衡网络资源和业务能耗之间的关系，在合理分配资源的情况下最大程度地减少云边协同网络的能耗，提高了云边协同网络的灵活性和可拓展性。
2.对每一个连接请求，根据任务的优先级别，运用OpenFlow扩展协议，由软件定义的中心控制器发出命令集，通过云边协同网络动态变化的拓扑 信息，采用K条最短路径算法计算业务到边缘计算服务器之间的工作路径。工作路径成功选择后，根据软件定义的中心控制器所掌握的资源占用状态信息，利用OpenFlow扩展协议，由软件定义的中心控制器下发资源分配命令，采用首次命中的频谱分配算法对路径进行频谱资源分配，需要同时满足频谱一致性和频谱连续性两个约束条件，然后对网络计算资源和频谱资源状态进行实时更新，提高了系统的灵活度。
本发明实施例还提供了一种软件定义云边协同网络能耗优化系统，包括：
集中控制模块，用于实时动态地了解与调整各个节点的资源信息，集中控制模块采用基于OpenFlow扩展协议的集中控管模式管理云边协同网络，建立软件定义云边协同网络；
应用服务需求模块，用于在软件定义云边协同网络中，根据应用服务需求，基于业务服务等级要求，接收由集中控制模块的集中控管模块发出命令集，建立工作路径与分配所需的频谱资源，完成应用服务需求的数据传输，更新计算资源并释放应用需求占用的网络资源。
本发明实施例的有益效果：
每一个连接请求成功建立后，根据用户发送功率、任务大小和光纤传输速率计算发射能耗，根据路径所经节点数和节点的负载情况计算路径节点能耗，以及根据计算边缘服务器的计算能力和边缘计算服务器处理数据单位时间所产生的能耗，计算出软件定义云边协同网络的能耗。通过软件定义云边协同网络能耗优化方法和系统，有效平衡网络资源和业务能耗之间的关系，在合理分配资源的情况下最大程度地减少云边协同网络的能耗，提高了云边协同网络的灵活性和可拓展性。
附图说明
通过参考附图会更加清楚的理解本发明的特征和优点，附图是示意性的而不应理解为对本发明进行任何限制，在附图中：
图1示出了本发明实施例中一种软件定义云边协同网络能耗优化方法 的流程图；
图2示出了本发明实施例中一种软件定义云边协同网络能耗优化方法的流程图；
图3示出了本发明实施例中一种软件定义云边协同网络能耗优化系统的网络模型图；
图4示出了本发明实施例中一种软件定义云边协同网络能耗优化系统的结构图。
具体实施方式
为使本发明实施例的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
实施例1
如图1所示，本发明实施例提供了一种软件定义云边协同网络能耗优化方法，包括：
步骤S10，获取连接请求集合。
本步骤包括：生成一组连接请求集合CR；其中，每一个连接请求CR(u,f,r)∈CR，u表示用户请求，f表示用户请求所需的频谱间隙数，r表示用户请求所需的计算资源数。
步骤S20，根据连接请求集合确定计算资源。
在本实施例中，对于每一个连接请求CR(u,f,r)，由软件定义的中心控制器判定用户请求所在的本地区域是否具有足够的计算资源，当本地区域内边缘计算服务器上的计算资源满足用户请求所需的计算资源时，由软件定义的中心控制器下发命令将用户请求发送到本地计算服务器进行处理，此时用户请求不经过网络中的其他节点，系统能耗只需考虑向本区域内边缘服务器的发送能耗和在本地区域内边缘服务器的计算能耗即可。若本地 区域计算资源不足，由软件定义的中心控制器做出决策，将计算迁移至边缘计算服务器或云计算服务器。
步骤S30，基于业务发送能耗、节点能耗和计算能耗通过K条最短路径算法获取最优工作路径。
在本实施例中，采用K条最短路径路算法计算从用户请求到边缘计算服务器之间的工作路径。K条最短路径路算法计算了K条候选路径作为路由选择，并按每条路径所产生的能耗由小到大进行排序，即能耗越小，优先选择权就越高。若排序在前的路径在某一段链路上发生阻塞，就依次选取排序在后的路径进行频谱资源分配，直到分配资源成功，或所有路径全都阻塞。
每一个连接请求成功建立后，根据用户发送功率、任务大小和光纤传输速率计算发射能耗，根据路径所经节点数和节点的负载情况计算路径节点能耗，以及根据计算边缘服务器的计算能力和边缘计算服务器处理数据单位时间所产生的能耗，计算出软件定义云边协同网络的能耗。通过软件定义云边协同网络能耗优化方法和系统，有效平衡网络资源和业务能耗之间的关系，在合理分配资源的情况下最大程度地减少云边协同网络的能耗，提高了云边协同网络的灵活性和可拓展性。
作为可选的实施方式，根据连接请求集合确定计算资源包括：
若本地区域计算资源满足连接请求集合，则由本地区域计算资源处理连接请求集合；
若本地区域计算资源不满足连接请求集合，则由本地区域计算资源和边缘区域计算资源处理连接请求集合；
若本地区域计算资源和边缘区域计算资源不满足连接请求集合，则由本地区域计算资源、边缘区域计算资源和云计算资源处理连接请求集合。
如图2所示，若本地边缘计算区域内的计算资源不足，基于软件定义的中心服务器判定边缘区域内服务器是否具有足够计算资源处理用户请求。若边缘区域服务器计算资源满足用户请求所需计算资源，软件定义的中心控制器下发命令至相邻服务器，需要把用户请求迁移到与本地区域内 的同一交换机上的其他区域边缘计算服务器上。此时，系统能耗包括用户请求向相邻服务器的发送能耗，途径交换机时所产生的节点能耗及相邻边缘服务器的计算能耗。
若软件定义的中央控制器检测到本地和相邻边缘计算区域服务器上的计算资源不能满足用户请求所需的计算资源，再判定云端服务器是否具有足够计算资源。若云端服务器具有足够计算资源，则用户请求通过交换机被发送到云端服务器上进行数据处理。此时，系统能耗包括用户向云端服务器发送请求的发送能耗，节点能耗及云端服务器的计算能耗。
作为可选的实施方式，基于业务发送能耗、节点能耗和计算能耗通过K条最短路径算法获取最优工作路径包括：
计算每一节点的负载任务数、节点负载和节点负载率。
在本实施例中，计算任一节点j的负载任务数X j(t)：
Figure PCTCN2021129564-appb-000013
计算任一节点j的负载Xc j(t)：
Figure PCTCN2021129564-appb-000014
计算任一节点j的负载率y j：
Figure PCTCN2021129564-appb-000015
其中，x l,j(t)表示时隙t由节点l到达节点j的任务数；
Figure PCTCN2021129564-appb-000016
表示节点j中第k个任务的大小；TR j表示节点j的资源总和；在云边协同网络G(U,B,I,S)中，B＝{1,2,…,b,…}表示一组基站；I＝{1,2,…,i,…}表示一组交换机；S＝B+I＝{1,2,…,s,…}表示一组边缘计算服务器可选取的位置。b、i、s均为大于等于1的自然数。
计算K条工作路径的发送能耗和节点能耗。
在本实施例中，计算发送能耗SEE：
Figure PCTCN2021129564-appb-000017
计算节点能耗NOE：
Figure PCTCN2021129564-appb-000018
其中，P u表示用户的发送功率；S u表示任务的大小；r表示光纤传输速率；J表示路径经过的节点数；
Figure PCTCN2021129564-appb-000019
表示第j个节点设备端口空载功耗；
Figure PCTCN2021129564-appb-000020
表示第j个节点设备端口满载功耗；y j表示第j个节点设备端口负载率。
根据用户请求所在层数获取计算能耗。
在本实施例中，若用户请求在边缘计算服务层，则计算能耗为：
Figure PCTCN2021129564-appb-000021
若用户请求在云计算服务层，则计算能耗为：
Figure PCTCN2021129564-appb-000022
其中，δ f表示边缘计算服务层的实时处理能力；
Figure PCTCN2021129564-appb-000023
表示边缘计算服务层处理数据单位时间所产生的能耗；δ c表示云计算服务层分配给每个任务的处理能力；
Figure PCTCN2021129564-appb-000024
表示云计算服务层处理数据单位时间所产生的能耗。
按照最低能耗对K条工作路径进行排序，得到多个备选路径；
在本实施例中，在得到多个备选路径之后，还包括：
基于频谱一致性和频谱连续性的约束条件，采用首次命中的频谱分配算法，根据备选路径上所有链路的频谱资源状态生成频谱资源表并进行编号；
从标号小的一端开始查找可用的频谱间隙；若当前频谱间隙可用，则进行频谱资源分配，并更新频谱状态；若未找到可用频谱，则频谱资源分配失败，业务存在阻塞。
按照能耗由低到高检测、排除存在阻塞的备选路径，得到最优工作路径，并进行频谱资源分配。
连接请求CR(u,f,r)成功建立工作路径后，告知基于软件定义的中央控制器可以进行用户请求的数据业务传输，根据频谱一致性和频谱连续性的约束条件对工作路径进行频谱资源分配。在具体实施例中，采用首次命中的频谱分配算法，根据路径上所有链路的频谱资源状态生成一张频谱资源表进行编号，从标号小的一端开始查找可用的频谱间隙。如果找到可用的频谱间隙则进行频谱资源分配并进行频谱状态更新；如果没有找到则频谱分配失败，业务阻塞。
在本实施例中，在得到最优工作路径，并进行频谱资源分配之后，还包括：对边缘区域计算资源进行更新，记录成功建立连接请求数。
对每一个连接请求，根据任务的优先级别，运用OpenFlow扩展协议，由软件定义的中心控制器发出命令集，通过云边协同网络动态变化的拓扑信息，采用K条最短路径算法计算业务到边缘计算服务器之间的工作路径。工作路径成功选择后，根据软件定义的中心控制器所掌握的资源占用状态信息，利用OpenFlow扩展协议，由软件定义的中心控制器下发资源分配命令，采用首次命中的频谱分配算法对路径进行频谱资源分配，需要同时满足频谱一致性和频谱连续性两个约束条件，然后对网络计算资源和频谱资源状态进行实时更新，提高了系统的灵活度。
实施例2
本发明实施例提供了一种软件定义云边协同网络能耗优化系统，包括集中控制模块和应用服务需求模块，其中：集中控制模块用于实时动态地了解与调整各个节点的资源信息，集中控制模块采用基于OpenFlow扩展协议的集中控管模式管理云边协同网络，建立软件定义云边协同网络；应用服务需求模块用于在软件定义云边协同网络中，根据应用服务需求，基于业务服务等级要求，接收由集中控制模块的集中控管模块发出命令集，建立工作路径与分配所需的频谱资源，完成应用服务需求的数据传输，更新计算资源并释放应用需求占用的网络资源。
如图3所示，集中控制模块主要包含：
—基于OpenFlow扩展协议的集中管控模块，该模块主要获得全部云边协同网络的资源信息与网络拓扑信息，监控每一个节点的状态，能够根据业务需求计算传输路径，为应用服务需求分配与释放相应的网络资源。
—云边协同网络初始化模块：通过OpenFlow扩展协议对云边协同网络进行初始化，在云边协同网络G(U,B,I,S)中，配置网络的拓扑信息、光网络连接状态、用户请求数、边缘计算服务器的数目、基站和交换机的数目。
—网络可编程硬件模块，该模块主要功能是引入了可编程的硬件模块，应用于云边协同网络控制状态的监控，感知网络资源使用状态。
—嵌入软件开发模块，该模块能够把开发的软件程序进行嵌入操作，并能够根据云边协同网络需要进行动态修改，同时也可以根据业务需求变化状态，修改软件的状态参数等信息。
—网络状态监控模块：主要完成对频谱灵活光网络初始化、连接请求生成、业务优先级选择、边缘计算服务器选择、工作路径建立、频谱资源分配、计算资源更新、资源释放、网络能耗计算的状态监控功能，以实现在计算资源分配时尽可能减少系统能耗的目标。
—判决和预警模块：执行各个模块之间的协调功能，以及每个模块是否建立成功的判决与预警功能，完成移动边缘计算中减少系统能耗的目标。
应用服务需求模块主要包含：
—应用服务需求模块：该模块主要是用户根据需求发出的服务请求，配置连接请求数目、不同连接请求所需的频谱间隙个数和计算资源等信息。
—软件定义的中心控制器的命令集模块：该模块能够根据云边协同网络的资源信息状态，以及执行在集中控制模块中基于OpenFlow扩展协议的集中控管模块的命令集合，完成应用需求的分发机制。首先，判定用户请求的本地服务器是否具有用户请求所需计算资源，若本地服务器计算资源足够则用户请求直接在本地进行处理。若本地服务器计算资源不够，考虑本地区域外的其他区域边缘计算服务器是否具有用户请求所需的计算资 源，若其他区域的服务器具有足够的计算资源，则将用户请求通过交换机迁移到其他区域进行处理。若本地和其他区域的计算资源都不满足用户请求的计算资源，则将用户请求通过交换机迁移到云服务器进行处理。
—工作路径建立模块：根据连接请求CR(u,f,r)的用户请求和处理请求的边缘计算服务器，采用K条最短路径算法，计算出从用户请求到服务器的K条候选路径，以便查找出最优的路径作为工作路径。
—业务优先级选择模块：根据K条最短路径算法，计算每条工作路径的用户发送能耗，路径节点能耗延和计算能耗总和，选择能耗低的工作路径进行业务处理传输和处理。
—频谱资源分配模块：根据连接请求CR(u,f,r)所需的频谱间隙数f，在所选择的工作路径中查找满足连接请求所需的带宽资源，若同时满足频谱连续性与频谱一致性双重约束条件，则成功建立连接请求；若不能同时满足频谱连续性与频谱一致性双重约束条件，则连接请求建立失败。
—计算资源更新模块：在频谱资源成功分配后，对处理用户请求的边缘计算服务器进行计算资源的实时更新。
—资源释放模块：在连接请求成功传输后，对工作路径占用的频谱资源进行资源释放。同时，对处理用户请求的边缘计算服务器的计算资源进行释放。最后，将连接请求建立的工作路径进行信息清除。
—能耗计算模块：在每一个连接请求成功建立请求后，记录每一个连接请求传输路径所经过的节点信息，根据实施例1中公式的计算每一个连接请求的用户发送能耗及节点能耗，再记录用户请求确定的边缘计算服务器的位置，计算每个连接请求的计算能耗。
实施例3
如图4所示，层1有三个本地区域，每个区域内都有基站和对应的边缘计算服务器，层2是由交换机和对应的边缘计算服务器组成的云端网络。
首先，建立软件定义云边协同网络。具体描述如下：
(1)根据云边协同网络的可编程特点，基于软件嵌入网络节点模式，融合OpenFlow代理商的编程网络节点，例如图中的可编程节点1、2、3、 4、5、6、7、8、9；
(2)根据节点间相互连接的特点，通过采用OpenFlow扩展协议的传输平面与控制平面相分离原则，由软件定义的中心控制器控管每个节点，建立软件定义云边协同网络。
其次，建立业务服务需求。具体描述如下：
(1)通过OpenFlow扩展协议对云边协同网络进行初始化，获取网络的资源信息和网络拓扑信息；
(2)通过基站发出用户请求，如图3①表示；
(3)通过中心控制器，可以感知用户请求，判断对应用户请求的处理位置，通过OpenFlow扩展协议的发出命令集，如图3②表示，假设需要在节点1到3之间建立传输路径；
(4)通过OpenFlow扩展协议的中心控制器来计算满足用户请求的K条路径，根据实施例1中的公式分别计算K条路径的发送能耗、节点能耗和计算能耗总和，并将K条路径的能耗和路径负载信息反馈到中央控制器中，最终先择负载和能耗相对最优的路径作为用户请求的工作路径1-3，如图3③表示；
(5)由可编程节点A和C发现响应命令，告知中心控制器，如图3④表示，用户请求的路径建立成功，可以进行服务需求的数据业务传输，完成数据传输后，释放资源并将连接请求建立的路径进行信息清除；
(6)通过OpenFlow扩展协议对云边协同网络的网络资源信息和网络拓扑信息进行更新。
为了建立软件定义云边协同网络，承载用户请求的数据传输，保证用户请求在软件定义云边协同网络中安全高效运行，其核心在于网络节点引入可编程功能，通过基于OpenFlow扩展协议的集中控制器，可以对整个软件定义云边协同网络进行实时状态监控。根据应用服务的属性需求，建立云边协同网络的传输路径，完成业务需求的端到端传输。
为了快速完成云边协同网络与统一云边协同网络资源调度功能，增加云边协同网络节点的可编程性，通过软件接入的方式进行动态管理，保证 云边协同网络功能与资源能够通过OpenFlow扩展协议进行集中控制与管理，简化云边协同网络的资源调度与优化问题，实现软件定义云边协同网络能耗优化方法。
本领域技术人员可以理解，实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，所述存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory，ROM)、随机存储记忆体(Random Access Memory，RAM)、快闪存储器(Flash Memory)、硬盘(Hard Disk Drive，缩写：HDD)或固态硬盘(Solid-State Drive，SSD)等；所述存储介质还可以包括上述种类的存储器的组合。
虽然结合附图描述了本发明的实施例，但是本领域技术人员可以在不脱离本发明的精神和范围的情况下作出各种修改和变型，这样的修改和变型均落入由所附权利要求所限定的范围之内。

Claims (10)

  1. 一种软件定义云边协同网络能耗优化方法，其特征在于，包括：
    获取连接请求集合；
    根据所述连接请求集合确定计算资源；
    基于业务发送能耗、节点能耗和计算能耗通过K条最短路径算法获取最优工作路径。
  2. 根据权利要求1所述的软件定义云边协同网络能耗优化方法，其特征在于，获取连接请求集合包括：
    生成一组连接请求集合CR；其中，每一个连接请求CR(u,f,r)∈CR，u表示用户请求，f表示用户请求所需的频谱间隙数，r表示用户请求所需的计算资源数。
  3. 根据权利要求2所述的软件定义云边协同网络能耗优化方法，其特征在于，根据所述连接请求集合确定计算资源包括：
    若本地区域计算资源满足所述连接请求集合，则由所述本地区域计算资源处理所述连接请求集合；
    若本地区域计算资源不满足所述连接请求集合，则由所述本地区域计算资源和边缘区域计算资源处理所述连接请求集合；
    若本地区域计算资源和所述边缘区域计算资源不满足所述连接请求集合，则由所述本地区域计算资源、所述边缘区域计算资源和云计算资源处理所述连接请求集合。
  4. 根据权利要求3所述的软件定义云边协同网络能耗优化方法，其特征在于，基于业务发送能耗、节点能耗和计算能耗通过K条最短路径算法获取最优工作路径包括：
    计算每一节点的负载任务数、节点负载和节点负载率；
    计算K条工作路径的发送能耗和节点能耗；
    根据所述用户请求所在层数获取计算能耗；
    按照最低能耗对K条所述工作路径进行排序，得到多个备选路径；
    按照能耗由低到高检测、排除存在阻塞的所述备选路径，得到所述最 优工作路径，并进行频谱资源分配。
  5. 根据权利要求4所述的软件定义云边协同网络能耗优化方法，其特征在于，计算每一节点的负载任务数、节点负载和节点负载率包括：
    计算任一节点j的负载任务数X j(t)：
    Figure PCTCN2021129564-appb-100001
    计算任一所述节点j的负载Xc j(t)：
    Figure PCTCN2021129564-appb-100002
    计算任一所述节点j的负载率y j：
    Figure PCTCN2021129564-appb-100003
    其中，x l,j(t)表示时隙t由节点l到达所述节点j的任务数；
    Figure PCTCN2021129564-appb-100004
    表示所述节点j中第k个任务的大小；TR j表示所述节点j的资源总和；B＝{1,2,…,b,…}表示一组基站；I＝{1,2,…,i,…}表示一组交换机；S＝B+I＝{1,2,…,s,…}表示一组边缘计算服务器可选取的位置。
  6. 根据权利要求4所述的软件定义云边协同网络能耗优化方法，其特征在于，计算K条工作路径的发送能耗和节点能耗包括：
    计算发送能耗SEE：
    Figure PCTCN2021129564-appb-100005
    计算节点能耗NOE：
    Figure PCTCN2021129564-appb-100006
    其中，P u表示用户的发送功率；S u表示任务的大小；r表示光纤传输速率；J表示路径经过的节点数；
    Figure PCTCN2021129564-appb-100007
    表示第j个节点设备端口空载功耗；
    Figure PCTCN2021129564-appb-100008
    表示第j个节点设备端口满载功耗；y j表示第j个节点设备端口负载率。
  7. 根据权利要求4所述的软件定义云边协同网络能耗优化方法，其特征在于，根据所述用户请求所在层数获取计算能耗包括：
    若所述用户请求在边缘计算服务层，则计算能耗为：
    Figure PCTCN2021129564-appb-100009
    若所述用户请求在云计算服务层，则计算能耗为：
    Figure PCTCN2021129564-appb-100010
    其中，δ f表示边缘计算服务层的实时处理能力；
    Figure PCTCN2021129564-appb-100011
    表示边缘计算服务层处理数据单位时间所产生的能耗；δ c表示云计算服务层分配给每个任务的处理能力；
    Figure PCTCN2021129564-appb-100012
    表示云计算服务层处理数据单位时间所产生的能耗。
  8. 根据权利要求4所述的软件定义云边协同网络能耗优化方法，其特征在于，在得到多个备选路径之后，还包括：
    基于频谱一致性和频谱连续性的约束条件，采用首次命中的频谱分配算法，根据所述备选路径上所有链路的频谱资源状态生成频谱资源表并进行编号；
    从标号小的一端开始查找可用的频谱间隙；若当前频谱间隙可用，则进行频谱资源分配，并更新频谱状态；若未找到可用频谱，则频谱资源分配失败，业务存在阻塞。
  9. 根据权利要求8所述的软件定义云边协同网络能耗优化方法，其特征在于，在得到所述最优工作路径，并进行频谱资源分配之后，还包括：
    对所述边缘区域计算资源进行更新，记录成功建立连接请求数。
  10. 一种软件定义云边协同网络能耗优化系统，其特征在于，包括：
    集中控制模块，用于实时动态地了解与调整各个节点的资源信息，所述集中控制模块采用基于OpenFlow扩展协议的集中控管模式管理云边协同网络，建立软件定义云边协同网络；
    应用服务需求模块，用于在软件定义云边协同网络中，根据应用服务需求，基于业务服务等级要求，接收由所述集中控制模块的集中控管模块发出命令集，建立工作路径与分配所需的频谱资源，完成应用服务需求的数据传输，更新计算资源并释放应用需求占用的网络资源。
PCT/CN2021/129564 2021-06-01 2021-11-09 软件定义云边协同网络能耗优化方法和系统 WO2022252496A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110609432.2A CN113364850B (zh) 2021-06-01 2021-06-01 软件定义云边协同网络能耗优化方法和系统
CN202110609432.2 2021-06-01

Publications (1)

Publication Number Publication Date
WO2022252496A1 true WO2022252496A1 (zh) 2022-12-08

Family

ID=77530860

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/129564 WO2022252496A1 (zh) 2021-06-01 2021-11-09 软件定义云边协同网络能耗优化方法和系统

Country Status (2)

Country Link
CN (1) CN113364850B (zh)
WO (1) WO2022252496A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116389491A (zh) * 2023-04-03 2023-07-04 中国电子技术标准化研究院 一种云边算力资源自适应计算系统
CN116455748A (zh) * 2023-06-16 2023-07-18 深圳市友恺通信技术有限公司 一种应用于网络设备运维的人工智能监控系统及方法
CN116955996A (zh) * 2023-09-15 2023-10-27 北京光函数科技有限公司 一种基于云无线接入网的云边协同智能推理方法及系统
CN118158218A (zh) * 2024-05-09 2024-06-07 杭州应敏科技有限公司 一种基于物联网网关的通信方法及系统

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113364850B (zh) * 2021-06-01 2023-02-14 苏州路之遥科技股份有限公司 软件定义云边协同网络能耗优化方法和系统
CN113742046A (zh) * 2021-09-17 2021-12-03 苏州大学 流量疏导的云边计算网络计算资源均衡调度方法及系统
CN114116211B (zh) * 2021-11-22 2024-04-16 中通服咨询设计研究院有限公司 面向移动边缘计算的业务流量分流优化方法与系统
CN114363984B (zh) * 2021-12-16 2022-11-25 苏州大学 一种云边协同光载网络频谱资源分配方法及系统
CN117203944A (zh) * 2022-05-26 2023-12-08 亚信科技（中国）有限公司 算力网络的资源调度方法
CN115002811A (zh) * 2022-08-01 2022-09-02 阿里巴巴达摩院(杭州)科技有限公司 专网管控方法、数据处理方法、设备控制方法及设备

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105337861A (zh) * 2015-11-18 2016-02-17 东北大学 一种基于能效优先和认知理论的路由方法
EP3340534A1 (en) * 2016-12-20 2018-06-27 Alcatel Lucent A local sdn controller and corresponding method of performing network control and management functions
CN109245918A (zh) * 2018-08-20 2019-01-18 中天宽带技术有限公司 软件定义频谱灵活光网络能耗优化方法及其系统
CN109800072A (zh) * 2019-01-22 2019-05-24 深圳市简智联信息科技有限公司 基于边缘计算的任务调度优化方法和装置
CN110446122A (zh) * 2019-05-31 2019-11-12 广东电网有限责任公司 一种光纤无线融合网络跨域协同资源分配方法
CN111901424A (zh) * 2020-07-28 2020-11-06 苏州大学 云边协同网络资源平滑迁移与重构方法及系统
AU2021101131A4 (en) * 2021-03-03 2021-05-06 Jeevan Bhatia Qos‑aware optical fog‑assisted cyber‑physical system in 5g ready heterogeneous network and method for the same
CN113364850A (zh) * 2021-06-01 2021-09-07 苏州路之遥科技股份有限公司 软件定义云边协同网络能耗优化方法和系统

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020023115A1 (en) * 2018-07-27 2020-01-30 Futurewei Technologies, Inc. Task offloading and routing in mobile edge cloud networks
EP3683677A1 (en) * 2019-01-15 2020-07-22 Siemens Aktiengesellschaft Method and allocator for allocating computation resources amongst a plurality of devices arranged in a network
CN110647391B (zh) * 2019-09-27 2022-04-12 北京邮电大学 面向星地协同网络的边缘计算方法及系统
CN112134916B (zh) * 2020-07-21 2021-06-11 南京邮电大学 一种基于深度强化学习的云边协同计算迁移方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105337861A (zh) * 2015-11-18 2016-02-17 东北大学 一种基于能效优先和认知理论的路由方法
EP3340534A1 (en) * 2016-12-20 2018-06-27 Alcatel Lucent A local sdn controller and corresponding method of performing network control and management functions
CN109245918A (zh) * 2018-08-20 2019-01-18 中天宽带技术有限公司 软件定义频谱灵活光网络能耗优化方法及其系统
CN109800072A (zh) * 2019-01-22 2019-05-24 深圳市简智联信息科技有限公司 基于边缘计算的任务调度优化方法和装置
CN110446122A (zh) * 2019-05-31 2019-11-12 广东电网有限责任公司 一种光纤无线融合网络跨域协同资源分配方法
CN111901424A (zh) * 2020-07-28 2020-11-06 苏州大学 云边协同网络资源平滑迁移与重构方法及系统
AU2021101131A4 (en) * 2021-03-03 2021-05-06 Jeevan Bhatia Qos‑aware optical fog‑assisted cyber‑physical system in 5g ready heterogeneous network and method for the same
CN113364850A (zh) * 2021-06-01 2021-09-07 苏州路之遥科技股份有限公司 软件定义云边协同网络能耗优化方法和系统

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116389491A (zh) * 2023-04-03 2023-07-04 中国电子技术标准化研究院 一种云边算力资源自适应计算系统
CN116389491B (zh) * 2023-04-03 2023-11-07 中国电子技术标准化研究院 一种云边算力资源自适应计算系统
CN116455748A (zh) * 2023-06-16 2023-07-18 深圳市友恺通信技术有限公司 一种应用于网络设备运维的人工智能监控系统及方法
CN116455748B (zh) * 2023-06-16 2023-08-25 深圳市友恺通信技术有限公司 一种应用于网络设备运维的人工智能监控系统及方法
CN116955996A (zh) * 2023-09-15 2023-10-27 北京光函数科技有限公司 一种基于云无线接入网的云边协同智能推理方法及系统
CN116955996B (zh) * 2023-09-15 2023-12-05 北京光函数科技有限公司 一种基于云无线接入网的云边协同智能推理方法及系统
CN118158218A (zh) * 2024-05-09 2024-06-07 杭州应敏科技有限公司 一种基于物联网网关的通信方法及系统

Also Published As

Publication number Publication date
CN113364850B (zh) 2023-02-14
CN113364850A (zh) 2021-09-07

Similar Documents

Publication Publication Date Title
WO2022252496A1 (zh) 软件定义云边协同网络能耗优化方法和系统
WO2022021176A1 (zh) 云边协同网络资源平滑迁移与重构方法及系统
US9143452B2 (en) Data processing
US20170220563A1 (en) Method, computing device, and distributed file system for placement of file blocks within a distributed file system
EP3296870A1 (en) Cdn-based content management system
WO2023039965A1 (zh) 流量疏导的云边计算网络计算资源均衡调度方法及系统
WO2023024219A1 (zh) 云边协同网络中时延和频谱占用联合优化方法及系统
CN110505539B (zh) 物理光网络虚拟化映射方法、装置、控制器及存储介质
Ibrahim et al. Heuristic resource allocation algorithm for controller placement in multi-control 5G based on SDN/NFV architecture
Bu et al. Enabling adaptive routing service customization via the integration of SDN and NFV
WO2023108718A1 (zh) 一种云边协同光载网络频谱资源分配方法及系统
WO2021120633A1 (zh) 一种负载均衡方法及相关设备
Xu et al. PDMA: Probabilistic service migration approach for delay‐aware and mobility‐aware mobile edge computing
Bourhim et al. Inter-container communication aware container placement in fog computing
Choudhury et al. Shareon: Shared resource dynamic container migration framework for real-time support in mobile edge clouds
Kafle et al. Adaptive virtual network slices for diverse IoT services
EP3280102B1 (en) Optical path allocation method and apparatus
WO2022016969A1 (zh) 一种数据处理方法及装置
Carrozzo et al. Large-scale SDN experiments in federated environments
CN117579550A (zh) 一种基于可编程交换机和容器状态信息卸载的路由方法
Wu Deep reinforcement learning based multi-layered traffic scheduling scheme in data center networks
US20220408353A1 (en) User plane function selection and hosting for real-time applications
US11310128B2 (en) Software-definable network service configuration method
Cheng Research on data center load balancing technology based on SDN
Alghamdi et al. Fog Network Area Management Model for Managing Fog-cloud Resources in IoT Environment

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21943848

Country of ref document: EP

Kind code of ref document: A1