Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456
WO2024244891A1 - 一种面向移动算力网络的分流方法及装置 - Google Patents
[go: Go Back, main page]

WO2024244891A1 - 一种面向移动算力网络的分流方法及装置 - Google Patents

一种面向移动算力网络的分流方法及装置 Download PDF

Info

Publication number
WO2024244891A1
WO2024244891A1 PCT/CN2024/090991 CN2024090991W WO2024244891A1 WO 2024244891 A1 WO2024244891 A1 WO 2024244891A1 CN 2024090991 W CN2024090991 W CN 2024090991W WO 2024244891 A1 WO2024244891 A1 WO 2024244891A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
base station
information
node
computing
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.)
Ceased
Application number
PCT/CN2024/090991
Other languages
English (en)
French (fr)
Inventor
郭凤仙
闫实
郭鲁杰
彭木根
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing University of Posts and Telecommunications
Original Assignee
Beijing University of Posts and Telecommunications
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 Beijing University of Posts and Telecommunications filed Critical Beijing University of Posts and Telecommunications
Publication of WO2024244891A1 publication Critical patent/WO2024244891A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • 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 present disclosure relates to the wireless and computing fields, and in particular to a traffic diversion method, apparatus, computer equipment, computer-readable storage medium, computer program product, and computer program for a mobile computing network.
  • cloud computing provides nearly unlimited elastic computing power, but due to the backhaul latency, it is only suitable for computing tasks that require a lot of computing power but have low latency requirements, such as large-scale AI training.
  • edge computing places computing power at the edge of the network, greatly reducing transmission latency, and is suitable for latency-sensitive computing tasks that do not require high computing power, such as target recognition in unmanned driving and video compression tasks in virtual reality.
  • wireless access networks have introduced concepts such as virtualization, cloud computing, and intelligence, and have gradually become an infrastructure for general computing integration and an ideal edge computing platform.
  • a computing power network which aims to coordinate cloud-edge-end computing power through networking means, achieve the adaptation of computing tasks and heterogeneous computing power, meet the diverse needs of users, and improve computing power utilization.
  • the current computing power network focuses on computer networks and does not incorporate the wireless computing power brought by the cloudification of access networks into its computing power system.
  • the wireless side has a large number of open high-quality computing resources, but since the base station cannot receive non-UPF data input, the existing diversion technology solutions need to be forwarded by the UPF of the core network before they can access the local server. Although the UPF sinking solution reduces the physical distance between the UPF and the base station, there is still the problem of UPF detour, and the edge UPF The deployment cost is high.
  • the present disclosure aims to solve one of the technical problems in the related art at least to some extent.
  • the purpose of the present disclosure is to propose a traffic diversion method, device, computer equipment, computer-readable storage medium, computer program product and computer program for a mobile computing power network, which are used for ubiquitous multi-level computing power in a collaborative network.
  • the first embodiment of the present disclosure proposes a traffic distribution method for a mobile computing network, including:
  • Identify the service in the service request by the first base station search the routing table, and determine whether a matching item is found for the service;
  • the first base station maintains information about neighboring nodes and optional computing nodes, calculates decision parameters based on the maintained information and the service decision value, and diverts the service to the first base station local computing node, the second base station computing node, the edge computing node or the central computing node based on the decision parameters.
  • a traffic distribution method for a mobile computing network may also have the following additional technical features:
  • the user terminal accesses the wireless network through a first base station, and a traffic diversion device is embedded in the first base station to provide a traffic diversion function for the service initiated by the user terminal.
  • the following steps are included:
  • the destination IP address of the service is replaced with the target computing node IP address, the routing table information is maintained, and the service is forwarded.
  • obtaining the service decision value according to the wireless delay information and the service demand information includes:
  • the method of calculating and obtaining the decision parameter based on the maintained information includes:
  • the method further comprises:
  • the destination computing node When the destination computing node is a local server, forwarding is performed directly; when the destination node is a server near a base station, forwarding is performed to the near base station through a wireless network; when the destination computing node is an edge computing node or a central computing node, forwarding is performed through UPF.
  • the second aspect of the present disclosure proposes a traffic distribution device for a mobile computing network, comprising the following modules:
  • a sending module configured to send a service request to the first base station through a user terminal
  • a matching module configured to identify the service in the service request through the first base station, search a routing table, and determine whether a matching item is found for the service;
  • an identification module configured to determine business requirement information according to the business type if no matching item is found for the business
  • a detection module configured to obtain a service decision value according to wireless delay information and service demand information, wherein the wireless delay information is obtained by performing wireless delay detection on the service by the first base station;
  • the diversion module is used to maintain information of adjacent nodes and optional computing power nodes through the first base station, calculate the decision parameters according to the maintained information and the service decision value, and divert the service to the local computing power node of the first base station, the computing power node of the second base station, the edge computing power node or the central computing power node according to the decision parameters.
  • the apparatus further comprises a forwarding module, configured to:
  • the destination IP address of the service is replaced with the target computing node IP address, the routing table information is maintained, and the service is forwarded.
  • an embodiment of the third aspect of the present disclosure proposes a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor.
  • the processor executes the computer program, a traffic diversion method for a mobile computing power network as described in any embodiment of the first aspect above is implemented.
  • the fourth aspect of the present disclosure provides a computer-readable storage medium having stored thereon A computer program, which, when executed by a processor, implements a traffic diversion method for a mobile computing power network as described in any embodiment of the first aspect above.
  • the fifth aspect of the present disclosure proposes a computer program product, including a computer program, which, when executed by a processor, implements a traffic diversion method for a mobile computing power network as described in any embodiment of the first aspect above.
  • the sixth aspect of the present disclosure proposes a computer program, including a computer program code.
  • the computer program code When the computer program code is run on a computer, the computer executes a traffic diversion method for a mobile computing power network as described in any embodiment of the first aspect above.
  • the diversion method for mobile computing power networks proposed in the embodiments of the present disclosure utilizes the advantage of the wireless side's natural perception of business information, deploys a local diversion device inside the base station, and implements dynamic diversion based on business status, which can minimize the overhead caused by information perception and improve business service quality.
  • the target server for diversion according to business needs, so that the business is directly forwarded to the local server on the wireless side without passing through the core network, the computing power resources can be truly physically and logically sunk, and the computing power resources on the wireless side can be fully utilized to give play to its low latency advantage.
  • the use of local server resources is minimized to ensure that when applications with ultra-low latency requirements are accessed, there are local computing power resources to provide services for them.
  • FIG1 is a flow chart of a traffic diversion method for a mobile computing network provided in an embodiment of the present disclosure.
  • FIG2 is a schematic diagram of a technical route of a traffic diversion method for a mobile computing power network provided in an embodiment of the present disclosure.
  • FIG3 is a schematic diagram of the composition of a mobile computing power network offload system provided in an embodiment of the present disclosure.
  • FIG4 is a schematic diagram of a processing flow when a base station receives downlink data from a server provided by an embodiment of the present disclosure.
  • FIG5 is a flow chart of a traffic diversion device for a mobile computing network provided in an embodiment of the present disclosure.
  • FIG1 is a flow chart of a method for offloading traffic to a mobile computing network provided by an embodiment of the first aspect of the present disclosure. picture.
  • the traffic diversion method for the mobile computing network includes the following steps S101 to S105 .
  • S101 Sending a service request to a first base station through a user terminal.
  • the user terminal accesses the wireless network through the first base station, and the offloading device is embedded in the first base station to provide an offloading function for the service initiated by the user terminal.
  • the user terminal sends a service request to the base station.
  • the service request can be initiated through a URL or an anycast IP address of the service, and the service request may also include but is not limited to source IP, user identifier (such as IMSI, GUTI), service identifier, source IP and other information.
  • S102 Identify the service in the service request through the first base station, search the routing table, and determine whether a matching item is found for the service.
  • the following steps are included:
  • the destination IP address of the service is replaced with the target computing node IP address, the routing table information is maintained, and the service is forwarded.
  • S102 includes identifying the service request information, searching the routing table, and determining whether there is a matching item for the service. If yes, forwarding is performed directly, otherwise execution continues.
  • the matching information may include but is not limited to source IP address, destination IP address, user ID, service ID and other information, and the routing table information items may include but are not limited to the information in Table 1.
  • the access base station identifies the service information by identifying the service identifier or other means to determine the service requirement, such as the delay threshold Delay-threshold.
  • S104 Acquire a service decision value according to the wireless delay information and the service demand information, wherein the wireless delay information is obtained by performing wireless delay detection on the service by the first base station.
  • the base station can perform Delay-wireless detection on the service and maintain information about adjacent nodes and optional computing nodes.
  • Node information may include but is not limited to the node's IP address, deployed service information, wireless delay, computing node service capability (average delay for processing a single request), and other information.
  • obtaining a service decision value according to wireless delay information and service demand information includes:
  • the first base station maintains information about neighboring nodes and optional computing nodes, calculates decision parameters based on the maintained information and service decision values, and diverts services to the first base station local computing node, the second base station computing node, the edge computing node, or the central computing node based on the decision parameters.
  • the access base station calculates the decision parameters of the base station based on the detected information, and distributes the service to the local computing node, the second base station computing node, the edge computing node or the central computing node according to the decision parameters.
  • An example of a decision method is given below.
  • a method for calculating and obtaining a decision parameter based on the maintained information includes:
  • the method further comprises:
  • the destination computing node When the destination computing node is a local server, forwarding is performed directly; when the destination node is a server near a base station, forwarding is performed to the near base station through a wireless network; when the destination computing node is an edge computing node or a central computing node, forwarding is performed through UPF.
  • FIG2 is a schematic diagram of the technical route disclosed in the present invention.
  • FIG3 is a schematic diagram of the composition of the mobile computing network distribution system in the disclosed embodiment.
  • the user terminal is the initiator of the service
  • the user terminal accesses the wireless network through the first base station, and the offloading device is embedded therein to provide offloading function for the service initiated by the user terminal;
  • the shareable computing power is the general computing power or intelligent computing power of the base station cloud, which can provide computing power services for the base station's own functions and third-party services;
  • the second base station is a base station adjacent to the user terminal access base station and has a traffic distribution device and sharable computing power like the first base station;
  • Fig. 4 is a processing flow when the base station receives downlink data from the server.
  • the flow includes the following steps S402 to S403.
  • the base station receives a data packet from a server.
  • the data packet may come from a local computing node, a neighboring base station computing node, an edge computing node, or a central computing node.
  • the diversion method for mobile computing power networks proposed in the embodiments of the present disclosure utilizes the advantage of the wireless side's natural perception of business information, deploys a local diversion device inside the base station, and implements dynamic diversion based on business status, which can minimize the overhead caused by information perception and improve business service quality.
  • the target server for diversion according to business needs, so that the business is directly forwarded to the local server on the wireless side without passing through the core network, the computing power resources can be truly physically and logically sunk, and the computing power resources on the wireless side can be fully utilized to give play to its low latency advantage.
  • the use of local server resources is minimized to ensure that when applications with ultra-low latency requirements are accessed, there are local computing power resources to provide services for them.
  • the second aspect of the present disclosure proposes a traffic diversion device for a mobile computing network.
  • FIG5 is a schematic diagram of the structure of a traffic diversion device for a mobile computing network provided in an embodiment of the present disclosure.
  • the traffic distribution device for the mobile computing network includes: a sending module 100, a matching module 200, an identification module 300, a detection module 400, and a traffic distribution module 500, wherein:
  • a sending module configured to send a service request to the first base station through a user terminal
  • a matching module configured to identify the service in the service request through the first base station, search the routing table, and determine whether a matching item is found for the service;
  • An identification module used to determine business requirement information according to the business type if no matching item is found for the business
  • a detection module configured to obtain a service decision value according to wireless delay information and service demand information, wherein the wireless delay information is obtained by performing wireless delay detection on the service by the first base station;
  • the diversion module is used to maintain information about adjacent nodes and optional computing power nodes through the first base station, calculate decision parameters based on the maintained information and service decision values, and divert the service to the local computing power node of the first base station, the computing power node of the second base station, the edge computing power node or the central computing power node according to the decision parameters.
  • the apparatus further includes a forwarding module 600, configured to:
  • the destination IP address of the service is replaced with the IP address of the target computing node, the routing table information is maintained, and the service is forwarded.
  • an embodiment of the third aspect of the present disclosure proposes a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor.
  • the processor executes the computer program, the traffic diversion method for the mobile computing power network as described in any embodiment of the first aspect above is implemented.
  • an embodiment of the fourth aspect of the present disclosure proposes a computer-readable storage medium on which a computer program is stored.
  • the computer program When executed by a processor, it implements the traffic diversion method for a mobile computing power network as described in any embodiment of the first aspect above.
  • the fifth aspect of the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements the traffic diversion method for a mobile computing power network as described in any embodiment of the first aspect.
  • a sixth aspect of the present disclosure provides a computer program.
  • the computer program code runs on a computer
  • the computer executes the traffic diversion method for a mobile computing power network as described in any one of the embodiments of the first aspect.
  • first and second are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features.
  • a feature defined as “first” or “second” may explicitly or implicitly include at least one of the features.
  • “plurality” means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

Landscapes

  • Mobile Radio Communication Systems (AREA)

Abstract

公开了一种面向移动算力网络的分流方法、装置、计算机设备、计算机可读存储介质、计算机程序产品和计算机程序。该方法包括，通过用户终端向第一基站发送业务请求；通过第一基站识别业务请求中的业务，查找路由表，判断业务是否找到匹配项；若业务未找到匹配项，则根据业务类型确定业务需求信息；根据无线延迟信息和业务需求信息获取业务判决值，其中，无线延迟信息通过第一基站对业务进行无线延迟检测得到；通过第一基站维护有临近节点以及可选算力节点的信息，根据维护的信息以及业务判决值计算判决参数，根据判决参数将业务分流至第一基站本地算力节点、第二基站算力节点、边缘算力节点或是中心算力节点。

Description

一种面向移动算力网络的分流方法及装置
相关申请的交叉引用
本申请要求在2023年05月29日在中国提交的中国专利申请号2023106201037的优先权，其全部内容通过引用并入本文。
技术领域
本公开涉及无线、计算领域，具体涉及一种面向移动算力网络的分流方法、装置、计算机设备、计算机可读存储介质、计算机程序产品和计算机程序。
背景技术
随着万物互联的数字时代的到来，新业态、新场景与新模式不断涌现，推动了海量数据产生，使得各行各业对通信和计算提出了更为迫切的需求。为应对多样化业务对算力的需求，云计算、边缘计算和雾计算等计算技术不断涌现。其中，云计算提供近乎无限制的弹性算力，但是因为回程时延，仅适用于需求大量算力但是时延要求不高的计算任务，比如大规模AI训练。为解决云计算所面临的高时延问题，边缘计算将算力放在网络边缘，大大降低了传输时延，适用于算力要求不高的时延敏感型计算任务，如无人驾驶中的目标识别以及虚拟现实中的视频压缩任务。与此同时，为了提升通信性能，赋能移动边缘计算，无线接入网引入虚拟化、云化、智能化等概念，逐渐成为了通算融合的基础设施，成为了理想的边缘算力平台。
为了协同网络中的泛在多级算力，运营商提出了算力网络，旨在通过网络化手段协同云边端协同算力，实现计算任务与异构算力的适配，满足用户的多样化需求，改善算力利用率。但是当前算力网络进聚焦于计算机网络，并没有将接入网云化带来的无线算力纳入其算力体系。
为充分利用网络中的泛在算力，其关键在于计算任务与算力资源的匹配。具体而言，需要根据业务特性和算力服务能力，将业务分发至合适的算力节点。本地分流技术是在移动通信网络中实现上述愿景的有效技术手段。目前移动通信网络中的分流技术主要有两种方式，一种是UPF下沉方案，另一种是在基站和核心网之间部署边缘计算网关实现业务分流。现有技术主要由一下缺点：
1)无线侧具有大量开放的优质算力资源，但是由于基站无法接收非UPF的数据输入，现有的分流技术方案在数据面都需要经过核心网的UPF转发才能够接入本地服务器。UPF下沉方案虽然减小了UPF和基站的物理距离，但仍然存在着UPF绕路问题，并且边缘UPF 的部署成本较大。
2)要为业务提供Qos保障，需要充分考虑业务需求以及用户的状态信息，要实现精准决策还需要对通信状态和算力状态进行实时感知，然而现有技术方案无论是集中式还是分布式的感知，实现难度都较大，并且均会造成很大的信令开销。
3)现有的分流方案将所有业务按照统一的规则进行分流，缺乏灵活性。需要智能灵活的分流方案。
发明内容
本公开旨在至少在一定程度上解决相关技术中的技术问题之一。
为此，本公开的目的在于提出一种面向移动算力网络的分流方法、装置、计算机设备、计算机可读存储介质、计算机程序产品和计算机程序，用于协同网络中的泛在多级算力。
为达上述目的，本公开第一方面实施例提出了一种面向移动算力网络的分流方法，包括：
通过用户终端向第一基站发送业务请求；
通过所述第一基站识别所述业务请求中的业务，查找路由表，判断所述业务是否找到匹配项；
若所述业务未找到匹配项，则根据所述业务类型确定业务需求信息；
根据无线延迟信息和业务需求信息获取业务判决值，其中，所述无线延迟信息通过所述第一基站对所述业务进行无线延迟检测得到；
通过所述第一基站维护有临近节点以及可选算力节点的信息，根据维护的信息以及所述业务判决值计算判决参数，根据所述判决参数将所述业务分流至第一基站本地算力节点、第二基站算力节点、边缘算力节点或是中心算力节点。
另外，根据本公开上述实施例的一种面向移动算力网络的分流方法还可以具有以下附加的技术特征：
在一些实施例中，所述用户终端通过第一基站接入无线网络，分流装置嵌入所述第一基站，为所述用户终端发起的业务提供分流功能。
在一些实施例中，当判断所述业务找到匹配项节点或者根据所述判决参数得到分流节点后，包括：
将所述业务的目的IP地址替换成目标算力节点IP地址，维护路由表信息，并对所述业务进行转发。
在一些实施例中，所述根据无线延迟信息和业务需求信息获取业务判决值，包括：
计算业务判决值δ＝T_threshold-T_wireless，其中，T_threshold表示业务信息， T_wireless表示延迟信息。
在一些实施例中，所述根据维护的信息计算获得判决参数的方法，包括：
定义算力节点a、b、c、d的判决参数分别为Ta、Tb、Tc、Td，且Ta<Tb<Tc<Td；其中，所述判决参数根据该类业务历史的处理时延信息获得；
当Ta≤δ<Tb时，业务转发至算力节点a；
当Tb≤δ<Tc时，业务转发至算力节点b；
当Tc≤δ<Td时，业务转发至算力节点c；
当Td≤δ时，业务转发至算力节点d。
在一些实施例中，所述方法还包括：
当目的算力节点为本地服务器时，直接进行转发；当目的节点为临近基站的服务器时，通过无线网络转发到临近基站；当目的算力节点为边缘算力节点或中心算力节点时，通过UPF进行转发。
为达上述目的，本公开第二方面实施例提出了一种面向移动算力网络的分流装置，包括以下模块：
发送模块，用于通过用户终端向第一基站发送业务请求；
匹配模块，用于通过所述第一基站识别所述业务请求中的业务，查找路由表，判断所述业务是否找到匹配项；
识别模块，用于若所述业务未找到匹配项，则根据所述业务类型确定业务需求信息；
检测模块，用于根据无线延迟信息和业务需求信息获取业务判决值，其中，所述无线延迟信息通过所述第一基站对所述业务进行无线延迟检测得到；
分流模块，用于通过所述第一基站维护有临近节点以及可选算力节点的信息，根据维护的信息以及所述业务判决值计算判决参数，根据所述判决参数将所述业务分流至第一基站本地算力节点、第二基站算力节点、边缘算力节点或是中心算力节点。
在一些实施例中，所述装置还包括转发模块，用于：
将所述业务的目的IP地址替换成目标算力节点IP地址，维护路由表信息，并对所述业务进行转发。
为达上述目的，本公开第三方面实施例提出了一种计算机设备，包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时，实现如上第一方面任一实施例所述的一种面向移动算力网络的分流方法。
为达上述目的，本公开第四方面实施例提出了一种计算机可读存储介质，其上存储有 计算机程序，所述计算机程序被处理器执行时实现如上第一方面任一实施例所述的一种面向移动算力网络的分流方法。
为达上述目的，本公开第五方面实施例提出一种计算机程序产品，包括计算机程序，所述计算机程序在被处理器执行时实现如上第一方面任一实施例所述的一种面向移动算力网络的分流方法。
为达上述目的，本公开第六方面实施例提出一种计算机程序，包括计算机程序代码，当所述计算机程序代码在计算机上运行时，使得所述计算机执行如上第一方面任一实施例所述的一种面向移动算力网络的分流方法。
本公开实施例提出的面向移动算力网络的分流方法，利用无线侧天然感知业务信息的优势，在基站内部部署本地分流装置，实现基于业务状态的动态分流，能够将信息感知带来的开销降到最小，可以改善业务服务质量。根据业务需求选择分流的目标服务器，使业务不经过核心网直接转发到无线侧的本地服务器，可以真正的做到算力资源在物理和逻辑上的下沉，充分利用无线侧的算力资源，发挥其低时延优势。在保障业务需求的前提下，最大限度的减少本地服务器资源的使用，从而保证超低时延需求的应用接入时，有本地算力资源可以为其提供服务。
附图说明
本公开上述的和/或附加的方面和优点从下面结合附图对实施例的描述中将变得明显和容易理解，其中：
图1为本公开实施例所提供的一种面向移动算力网络的分流方法的流程示意图。
图2为本公开实施例所提供的一种面向移动算力网络的分流方法的技术路线示意图。
图3为本公开实施例所提供的一种移动算力网络分流系统组成示意图。
图4为本公开实施例所提供的一种基站收到来自服务器的下行数据时处理流程示意图。
图5为本公开实施例所提供的一种面向移动算力网络的分流装置的流程示意图。
具体实施方式
下面详细描述本公开的实施例，所述实施例的示例在附图中示出，其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附图描述的实施例是示例性的，旨在用于解释本公开，而不能理解为对本公开的限制。
下面参考附图描述本公开实施例的面向移动算力网络的分流方法。
图1为本公开第一方面实施例所提供的一种面向移动算力网络的分流方法的流程示意 图。
如图1所示，该面向移动算力网络的分流方法包括以下步骤S101至S105。
S101：通过用户终端向第一基站发送业务请求。
在一些实施例中，用户终端通过第一基站接入无线网络，分流装置嵌入第一基站，为用户终端发起的业务提供分流功能。
具体的，用户终端向基站发送业务请求。业务请求可以通过URL或者业务的任播IP地址发起，业务请求中还可能包含但不限于源IP、用户标识(如IMSI、GUTI)、业务标识、源IP等信息。
S102：通过第一基站识别业务请求中的业务，查找路由表，判断业务是否找到匹配项。
在一些实施例中，当判断业务找到匹配项节点或者根据判决参数得到分流节点后，包括：
将所述业务的目的IP地址替换成目标算力节点IP地址，维护路由表信息，并对业务进行转发。
具体的，S102包括识别业务请求信息，查找路由表，判断该业务是否存在匹配项。如果是则直接进行转发，否则继续执行。其中匹配信息可以包括但不限于源IP地址、目的IP地址、用户标识、服务标识等信息，路由表信息项可以包括但不限于表1信息。
表1
S103：若业务未找到匹配项，则根据业务类型确定业务需求信息。
接入基站通过识别业务标识或是其他方式识别业务信息，确定业务需求，如时延阈值Delay-threshold。
S104：根据无线延迟信息和业务需求信息获取业务判决值，其中，无线延迟信息通过第一基站对业务进行无线延迟检测得到。
基站可以对业务进行Delay-wireless检测，并且维护有临近节点以及可选算力节点的信息。节点信息可以包括但不限于节点的IP地址、部署的服务信息、无线时延、算力节点服务能力(处理单次请求的平均时延)等信息。
在一些实施例中，根据无线延迟信息和业务需求信息获取业务判决值，包括：
计算业务判决值δ＝T_threshold-T_wireless，其中，T_threshold表示业务信息，T_wireless表示延迟信息。
S105：通过第一基站维护有临近节点以及可选算力节点的信息，根据维护的信息以及业务判决值计算判决参数，根据判决参数将业务分流至第一基站本地算力节点、第二基站算力节点、边缘算力节点或是中心算力节点。
接入基站根据检测的信息，计算获得基站的判决参数，并根据判决参数将业务分流至本地算力节点、第二基站算力节点、边缘算力节点或是中心算力节点。以下给出一种判决方法示例。
在一些实施例中，根据维护的信息计算获得判决参数的方法，包括：
定义算力节点a、b、c、d的判决参数分别为Ta、Tb、Tc、Td，且Ta<Tb<Tc<Td；其中，判决参数根据该类业务历史的处理时延信息获得；
当Ta≤δ<Tb时，业务转发至算力节点a；
当Tb≤δ<Tc时，业务转发至算力节点b；
当Tc≤δ<Td时，业务转发至算力节点c；
当Td≤δ时，业务转发至算力节点d。
在一些实施例中，该方法还包括：
当目的算力节点为本地服务器时，直接进行转发；当目的节点为临近基站的服务器时，通过无线网络转发到临近基站；当目的算力节点为边缘算力节点或中心算力节点时，通过UPF进行转发。
以上为完整的面向移动算力网络的分流方法流程，图2为本公开的技术路线示意图。
图3为公开实施例中移动算力网络分流系统组成示意图。其中：
1)用户终端是业务的发起方；
2)用户终端通过第一基站接入无线网络，分流装置嵌入其中，为用户终端发起的业务提供分流功能；
3)在第一基站中，可共享算力为基站云化的通用化算力或者智能算力，可以为基站本身功能以及第三方业务提供算力服务；
4)第二基站是与用户终端接入基站相邻的基站，和第一基站一样拥有分流装置和可共享算力；
5)通过核心网的UPF，业务可以从无线接入网被转发到边缘算力节点或中心算力节点。
图4为基站收到来自服务器的下行数据时处理流程。流程包括以下步骤S402至S403。
S401:基站接收来自服务器的数据包。该数据包可以来自本地算力节点、临近基站算力节点、边缘算力节点或中心算力节点。
S402:获取下行数据包的信息，查询维护的路由表。
S403:若找到匹配项，则将数据包的源IP地址修改为业务请求时的目的IP地址，并根据用户标识，转发给对应的用户终端；若未找到匹配项则直接丢弃数据包。
本公开实施例提出的面向移动算力网络的分流方法，利用无线侧天然感知业务信息的优势，在基站内部部署本地分流装置，实现基于业务状态的动态分流，能够将信息感知带来的开销降到最小，可以改善业务服务质量。根据业务需求选择分流的目标服务器，使业务不经过核心网直接转发到无线侧的本地服务器，可以真正的做到算力资源在物理和逻辑上的下沉，充分利用无线侧的算力资源，发挥其低时延优势。在保障业务需求的前提下，最大限度的减少本地服务器资源的使用，从而保证超低时延需求的应用接入时，有本地算力资源可以为其提供服务。
为了实现上述实施例，本公开第二方面实施例提出面向移动算力网络的分流装置。
图5为本公开实施例提供的一种面向移动算力网络的分流装置的结构示意图。
如图5所示，该面向移动算力网络的分流装置包括：发送模块100，匹配模块200，识别模块300，检测模块400，分流模块500，其中，
发送模块，用于通过用户终端向第一基站发送业务请求；
匹配模块，用于通过第一基站识别业务请求中的业务，查找路由表，判断业务是否找到匹配项；
识别模块，用于若业务未找到匹配项，则根据业务类型确定业务需求信息；
检测模块，用于根据无线延迟信息和业务需求信息获取业务判决值，其中，无线延迟信息通过第一基站对业务进行无线延迟检测得到；
分流模块，用于通过第一基站维护有临近节点以及可选算力节点的信息，根据维护的信息以及业务判决值计算判决参数，根据判决参数将业务分流至第一基站本地算力节点、第二基站算力节点、边缘算力节点或是中心算力节点。
在一些实施例中，该装置还包括转发模块600，用于：
将业务的目的IP地址替换成目标算力节点IP地址，维护路由表信息，并对所述业务进行转发。
为达上述目的，本公开第三方面实施例提出了一种计算机设备，包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时，实现如上第一方面任一实施例所述的面向移动算力网络的分流方法。
为达上述目的，本公开第四方面实施例提出了一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现如上第一方面任一实施例所述的面向移动算力网络的分流方法。
本公开第五方面实施例提供一种计算机程序产品，包括计算机程序，所述计算机程序在被处理器执行时实现上述第一方面任一实施例所述的面向移动算力网络的分流方法。
本公开第六方面实施例提供一种计算机程序，当所述计算机程序代码在计算机上运行时，使得所述计算机执行上述第一方面任一实施例所述的面向移动算力网络的分流方法。
需要说明的是，前述实施例对面向移动算力网络的分流方法的解释说明也适用于本公开实施例的装置、可读存储介质、计算机设备、计算机程序产品和计算机程序，此处不再赘述。
本公开所有实施例均可以单独被执行，也可以与其它实施例相结合被执行，均视为本公开要求的保护范围。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本公开的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不必须针对的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任一个或多个实施例或示例中以合适的方式结合。此外，在不相互矛盾的情况下，本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。在本公开的描述中，“多个”的含义是至少两个，例如两个，三个等，除非另有明确具体的限定。
尽管上面已经示出和描述了本公开的实施例，可以理解的是,上述实施例是示例性的,不能理解为对本公开的限制,本领域的普通技术人员在本公开的范围内可以对上述实施例进行变化、修改、替换和变型。

Claims (12)

  1. 一种面向移动算力网络的分流方法，包括以下步骤：
    通过用户终端向第一基站发送业务请求；
    通过所述第一基站识别所述业务请求中的业务，查找路由表，判断所述业务是否找到匹配项；
    若所述业务未找到匹配项，则根据所述业务类型确定业务需求信息；
    根据无线延迟信息和业务需求信息获取业务判决值，其中，所述无线延迟信息通过所述第一基站对所述业务进行无线延迟检测得到；
    通过所述第一基站维护有临近节点以及可选算力节点的信息，根据维护的信息以及所述业务判决值计算判决参数，根据所述判决参数将所述业务分流至第一基站本地算力节点、第二基站算力节点、边缘算力节点或是中心算力节点。
  2. 根据权利要求1所述的方法，其中所述用户终端通过第一基站接入无线网络，分流装置嵌入所述第一基站，为所述用户终端发起的业务提供分流功能。
  3. 根据权利要求1或2所述的方法，其中当判断所述业务找到匹配项节点或者根据所述判决参数得到分流节点后，包括：
    将所述业务的目的IP地址替换成目标算力节点IP地址，维护路由表信息，并对所述业务进行转发。
  4. 根据权利要求1至3中任一项所述的方法，其中所述根据无线延迟信息和业务需求信息获取业务判决值，包括：
    计算业务判决值δ＝T_threshold-T_wireless，其中，T_threshold表示业务信息，T_wireless表示延迟信息。
  5. 根据权利要求1至4中任一项所述的方法，其中所述根据维护的信息计算获得判决参数的方法，包括：
    定义算力节点a、b、c、d的判决参数分别为Ta、Tb、Tc、Td，且Ta<Tb<Tc<Td；其中，所述判决参数根据该类业务历史的处理时延信息获得；
    当Ta≤δ<Tb时，业务转发至算力节点a；
    当Tb≤δ<Tc时，业务转发至算力节点b；
    当Tc≤δ<Td时，业务转发至算力节点c；
    当Td≤δ时，业务转发至算力节点d。
  6. 根据权利要求5所述的方法，还包括：
    当目的算力节点为本地服务器时，直接进行转发；当目的节点为临近基站的服务器时，通过无线网络转发到临近基站；当目的算力节点为边缘算力节点或中心算力节点时，通过UPF进行转发。
  7. 一种面向移动算力网络的分流装置，包括以下模块：
    发送模块，用于通过用户终端向第一基站发送业务请求；
    匹配模块，用于通过所述第一基站识别所述业务请求中的业务，查找路由表，判断所述业务是否找到匹配项；
    识别模块，用于若所述业务未找到匹配项，则根据所述业务类型确定业务需求信息；
    检测模块，用于根据无线延迟信息和业务需求信息获取业务判决值，其中，所述无线延迟信息通过所述第一基站对所述业务进行无线延迟检测得到；
    分流模块，用于通过所述第一基站维护有临近节点以及可选算力节点的信息，根据维护的信息以及所述业务判决值计算判决参数，根据所述判决参数将所述业务分流至第一基站本地算力节点、第二基站算力节点、边缘算力节点或是中心算力节点。
  8. 根据权利要求7所述的装置，还包括转发模块，用于：
    将所述业务的目的IP地址替换成目标算力节点IP地址，维护路由表信息，并对所述业务进行转发。
  9. 一种计算机设备，包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序，其中所述处理器执行所述计算机程序时，实现如权利要求1至6中任一项所述的面向移动算力网络的分流方法。
  10. 一种计算机可读存储介质，其上存储有计算机程序，其中所述计算机程序被处理器执行时实现如权利要求1至6中任一项所述的面向移动算力网络的分流方法。
  11. 一种计算机程序产品，包括计算机程序，所述计算机程序在被处理器执行时实现如权利要求1至6中任一项所述的面向移动算力网络的分流方法。
  12. 一种计算机程序，包括计算机程序代码，当所述计算机程序代码在计算机上运行时，使得所述计算机执行如权利要求1至6中任一项所述的面向移动算力网络的分流方法。
PCT/CN2024/090991 2023-05-29 2024-04-30 一种面向移动算力网络的分流方法及装置 Ceased WO2024244891A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202310620103.7 2023-05-29
CN202310620103.7A CN116761224A (zh) 2023-05-29 2023-05-29 一种面向移动算力网络的分流方法及装置

Publications (1)

Publication Number Publication Date
WO2024244891A1 true WO2024244891A1 (zh) 2024-12-05

Family

ID=87958103

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2024/090991 Ceased WO2024244891A1 (zh) 2023-05-29 2024-04-30 一种面向移动算力网络的分流方法及装置

Country Status (2)

Country Link
CN (1) CN116761224A (zh)
WO (1) WO2024244891A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119815313A (zh) * 2024-12-26 2025-04-11 中国电信股份有限公司技术创新中心 服务提供方法及装置、存储介质及电子设备

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116761224A (zh) * 2023-05-29 2023-09-15 北京邮电大学 一种面向移动算力网络的分流方法及装置
CN119363644A (zh) * 2024-10-17 2025-01-24 中国联合网络通信集团有限公司 无线网络数据转发方法、装置及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112003907A (zh) * 2020-08-04 2020-11-27 北京大学深圳研究生院 网络资源需求及其中算力需求的确定性转发方法
WO2022028418A1 (zh) * 2020-08-04 2022-02-10 中国移动通信有限公司研究院 算力处理的网络系统、业务处理方法及设备
WO2022143750A1 (zh) * 2020-12-31 2022-07-07 维沃移动通信有限公司 信息处理方法、装置、设备及存储介质
CN114867065A (zh) * 2022-05-18 2022-08-05 中国联合网络通信集团有限公司 基站算力负载均衡方法、设备及存储介质
CN115794407A (zh) * 2022-12-15 2023-03-14 中国电信股份有限公司 计算资源分配方法、装置、电子设备及非易失性存储介质
CN116761224A (zh) * 2023-05-29 2023-09-15 北京邮电大学 一种面向移动算力网络的分流方法及装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112003907A (zh) * 2020-08-04 2020-11-27 北京大学深圳研究生院 网络资源需求及其中算力需求的确定性转发方法
WO2022028418A1 (zh) * 2020-08-04 2022-02-10 中国移动通信有限公司研究院 算力处理的网络系统、业务处理方法及设备
WO2022143750A1 (zh) * 2020-12-31 2022-07-07 维沃移动通信有限公司 信息处理方法、装置、设备及存储介质
CN114867065A (zh) * 2022-05-18 2022-08-05 中国联合网络通信集团有限公司 基站算力负载均衡方法、设备及存储介质
CN115794407A (zh) * 2022-12-15 2023-03-14 中国电信股份有限公司 计算资源分配方法、装置、电子设备及非易失性存储介质
CN116761224A (zh) * 2023-05-29 2023-09-15 北京邮电大学 一种面向移动算力网络的分流方法及装置

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119815313A (zh) * 2024-12-26 2025-04-11 中国电信股份有限公司技术创新中心 服务提供方法及装置、存储介质及电子设备

Also Published As

Publication number Publication date
CN116761224A (zh) 2023-09-15

Similar Documents

Publication Publication Date Title
JP7521005B2 (ja) ユーザプレーン機能の決定方法、情報提供方法、機器及び媒体
EP3993347B1 (en) Application migration
CN109495929B (zh) 一种业务处理方法、移动边缘计算设备及网络设备
WO2024244891A1 (zh) 一种面向移动算力网络的分流方法及装置
CN113079541B (zh) 一种报告信息的发送方法、装置及系统
CN108632055B (zh) 网络控制的方法和装置以及通信系统
US20160100316A1 (en) Mobile application specific networks
CN114007204B (zh) 基于中继通信与直连通信的通信选择方法和设备
CN116074199B (zh) 6g空天地场景细粒度按需服务方法
US20230216775A1 (en) Packet forwarding method, forwarding indication information advertising method, advertisement packet advertising method, and device
CN104660507A (zh) 数据流转发路由的控制方法及装置
CN108235376B (zh) 一种用户面锚点选择方法及装置
WO2020224307A1 (zh) 一种通信方法以及相关通信设备
WO2021164428A1 (zh) 通信方法、装置及系统
CN114553886A (zh) 数据传输方法和通信装置
US10862858B2 (en) Information centric approach in achieving anycast in machine type communications
WO2023005745A1 (zh) 报文转发方法、装置及系统、计算机可读存储介质
CN109196895A (zh) 一种报文传输方法、装置及系统
CN112449382B (zh) 一种数据传输方法和服务器
Bellaj et al. On-demand Geographic Forwarding for producer mobility support in Named Data Networking
CN108617012B (zh) 一种建立连接的方法及装置
CN117956435A (zh) 一种无线算力处理方法、装置及设备
CN112449398B (zh) 一种路径选择方法和核心网设备
Manogaran et al. Replication based Crowd Sensing for Optimal Service Response in 5G Communications using Information-Centric Wireless Sensor Networks.
CN115915199A (zh) 一种5gc架构对应的方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 24814096

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE