WO2025237231A1 - 通信方法、装置、通信设备及可读存储介质 - Google Patents

通信方法、装置、通信设备及可读存储介质

Info

Publication number
WO2025237231A1
WO2025237231A1 PCT/CN2025/094224 CN2025094224W WO2025237231A1 WO 2025237231 A1 WO2025237231 A1 WO 2025237231A1 CN 2025094224 W CN2025094224 W CN 2025094224W WO 2025237231 A1 WO2025237231 A1 WO 2025237231A1
Authority
WO
WIPO (PCT)
Prior art keywords
computing power
computing
network function
node
requirement
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
Application number
PCT/CN2025/094224
Other languages
English (en)
French (fr)
Inventor
刘棠青
王丹
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.)
China Mobile Communications Group Co Ltd
Research Institute of China Mobile Communication Co Ltd
Original Assignee
China Mobile Communications Group Co Ltd
Research Institute of China Mobile Communication Co Ltd
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 China Mobile Communications Group Co Ltd, Research Institute of China Mobile Communication Co Ltd filed Critical China Mobile Communications Group Co Ltd
Publication of WO2025237231A1 publication Critical patent/WO2025237231A1/zh
Pending legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/10Protocols in which an application is distributed across nodes in the network
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/131Protocols for games, networked simulations or virtual reality
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/50Network services
    • H04L67/60Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W60/00Affiliation to network, e.g. registration; Terminating affiliation with the network, e.g. de-registration
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W68/00User notification, e.g. alerting and paging, for incoming communication, change of service or the like
    • H04W68/02Arrangements for increasing efficiency of notification or paging channel

Definitions

  • This disclosure belongs to the field of communication technology, and specifically relates to a communication method, apparatus, communication device and readable storage medium.
  • Immersive services require networks to transmit and process multimodal data, including audio, video, haptic feedback, and information from various sensors. This places stringent requirements on network performance metrics such as transmission rate, latency, and reliability, as well as computing power such as processor and storage capabilities.
  • network performance metrics such as transmission rate, latency, and reliability, as well as computing power such as processor and storage capabilities.
  • computing power such as processor and storage capabilities.
  • the user experience of immersive applications is poor, failing to fully realize the user's immersive experience, and exhibiting noticeable lag and stuttering. Therefore, improving the user experience is an urgent problem to be solved.
  • the purpose of this disclosure is to provide a communication method, apparatus, communication device, and readable storage medium to address the problem of how to improve service experience.
  • a communication method is provided, applied to a first network function, including:
  • a communication method for use in a second network function, including:
  • the second network function sends the first request to the first network function.
  • a communication method applied to a first device, comprising:
  • the first device sends a second data packet to the second network function; wherein the second data packet carries a second requirement, the second requirement including at least one of the following: second task number, calculation type, calculation accuracy, and estimated task duration.
  • a communication device for use in a first network function, comprising:
  • the first receiving module is used to receive the first request sent by the second network function.
  • a communication device for use in a second network function, comprising:
  • the fourth sending module is used to send the first request to the first network function.
  • a communication device applied to a first device, comprising:
  • the seventh sending module is used to send a second data packet to the second network function; wherein the second data packet carries a second requirement, the second requirement including at least one of the following: second task number, calculation type, calculation precision, and estimated task duration.
  • a communication device including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method described in the first aspect, or the steps of the method described in the second aspect, or the steps of the method described in the third aspect.
  • a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect, or the method described in the second aspect, or the method described in the third aspect.
  • a computer program product including computer instructions that, when executed by a processor, implement the steps of the method described in the first aspect, or the method described in the second aspect, or the method described in the third aspect.
  • the solution in this embodiment can notify the first network function of the first requirement (such as computing power requirement or computing requirement), and then select a matching computing power node to calculate the business data, thereby reducing business lag, jitter and other issues, and improving the business experience.
  • the first network function of the first requirement such as computing power requirement or computing requirement
  • Figure 1 is a flowchart of a communication method provided in an embodiment of this disclosure
  • Figure 2 is a flowchart of the active registration process of computing resources in an embodiment of this disclosure
  • Figure 3 is a flowchart of the paging registration process of computing resources in an embodiment of this disclosure
  • Figure 4 is a flowchart of the deregistration process of computing resources in an embodiment of this disclosure
  • FIG. 5 is a flowchart of another communication method provided in an embodiment of this disclosure.
  • FIG. 6 is a flowchart of another communication method provided in an embodiment of this disclosure.
  • Figure 7 is a flowchart of the process of carrying computing power requirements along the way in an embodiment of this disclosure.
  • Figure 11 is a schematic diagram of the structure of a communication device provided in an embodiment of this disclosure.
  • first,” “second,” etc. used in this disclosure and in the claims are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this disclosure can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first,” “second,” etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more.
  • “and/or” indicates at least one of the connected objects, and the character “/” generally indicates that the preceding and following objects are in an "or” relationship.
  • a computing node can be described as a computing node, and the two have the same meaning.
  • computing power task can be described as computing task, and the two have the same meaning.
  • computing power resources can be described as computing resources, and the two have the same meaning.
  • computing resources may include mobile computing resources.
  • Figure 1 is a flowchart of a communication method provided in an embodiment of this disclosure. The method is applied to a first network function. As shown in Figure 1, the method includes the following steps:
  • Step 11 The first network function receives the first request sent by the second network function.
  • the first requirement may be a computing power requirement or a computing requirement.
  • a computing power requirement can be expressed as a computing requirement; both have the same meaning.
  • the first network function can be any of the following: Computing Control Function (CCF), Session Management Function (SMF), or computing network element.
  • CCF Computing Control Function
  • SMF Session Management Function
  • the first network function can also be a combined network function (NF) of CCF, SMF, and/or computing network element, or other NFs capable of performing similar functions; this embodiment does not limit this to any particular function.
  • NF network function
  • the second network function can be any of the following: User Plane Function (UPF) or computing node.
  • UPF User Plane Function
  • the second network function can also be a combined UPF and computing node NF, or other NFs capable of achieving similar functionality; this embodiment does not limit this to any particular type.
  • the first requirement may include, but is not limited to, at least one of the following:
  • the first task number is a task number used within the network to identify computing tasks, such as the identifier used within the 5th Generation Mobile Communication Technology Core (5GC).
  • 5GC 5th Generation Mobile Communication Technology Core
  • Terminal location information which can be selected as Tracking Area Identity (TAI) or Geographic Area, etc.
  • TAI Tracking Area Identity
  • Geographic Area etc.
  • the computing type can be, but is not limited to, an Accelerated Processing Unit (APU), a Central Processing Unit (CPU), a Data Processing Unit (DPU), a Graphics Processing Unit (GPU), a Neural Processing Unit (NPU), a Tensor Processing Unit (TPU), etc.
  • APU Accelerated Processing Unit
  • CPU Central Processing Unit
  • DPU Data Processing Unit
  • GPU Graphics Processing Unit
  • NPU Neural Processing Unit
  • TPU Tensor Processing Unit
  • the computational precision can be selected as, but is not limited to, Million Instructions Per Second (MIPS), DMIPS (Dhrystone MIPS, which represents the relative performance of the CPU when performing integer operations (Dhrystone), Operations Per Second (OPS), Floating Point Operations Per Second (FLOPS), etc.
  • MIPS Million Instructions Per Second
  • DMIPS Dynamic Metal Integrity Polymer
  • OPS Operations Per Second
  • FLOPS Floating Point Operations Per Second
  • the solution in this embodiment can notify the first network function of the first requirement (such as computing power requirement or computing requirement), and then select a matching computing power node to calculate the business data, thereby reducing business lag, jitter and other issues, and improving the business experience.
  • the first network function of the first requirement such as computing power requirement or computing requirement
  • the communication method may further include:
  • the first network function matches the first requirement with registered computing resources to obtain a first computing node.
  • This first computing node can be understood as either a matching computing node used to execute the computing task corresponding to the first task number in the first requirement, or a computing node matched with the computing task corresponding to the first task number in the first requirement. This allows for the selection of a suitable computing node to execute the computing task.
  • the matching information may include at least one of the following:
  • Types of computing resources such as APU, CPU, DPU, GPU, NPU, TPU, etc.;
  • NF ID Network Function Identifier
  • FQDN Fully Qualified Domain Name
  • GPSI Global Product and Service Identifier
  • Location information such as the location information of computing nodes
  • Terminal location such as TAI or Geographic Area
  • NF type such as UPF, Maintenance Endpoint (MEP), Vehicle-to-Everything server (V2X server), Third-party Application ( 3rd APP), User Equipment (UE), etc.;
  • MEP Maintenance Endpoint
  • V2X server Vehicle-to-Everything server
  • 3rd APP Third-party Application
  • UE User Equipment
  • the matching rules may include, but are not limited to, at least one of the following:
  • Prioritize matching computing power nodes that are closer to the second network function for example, under the same conditions of computing type and computing accuracy, prioritize matching computing power nodes that are closer to the second network function (such as UPF);
  • the second value can be pre-configured by the operator according to actual needs, such as N km; for example, if the distance between the computing resources of a certain computing node and the second network function (such as UPF) is greater than N km, then the computing node is considered to be unsatisfactory.
  • the computing task is divided into multiple sub-time periods and the tasks in the multiple sub-time periods are matched again. For example, if the estimated task time cannot be matched, the computing task can be divided into M sub-time periods and matched again, where M is an integer greater than 1.
  • the communication method in this disclosure embodiment may further include at least one of the following:
  • the first network function receives the first data packet
  • the first network function parses the first task number carried in the first data packet
  • the first network function forwards the first data packet to the first computing node according to the first task number.
  • the first network function (such as CCF) can receive the first data packet from UPF; the first data packet can be understood as a service data packet, which carries a first task number (i.e., network internal task number) for identifying the computing task.
  • a first task number i.e., network internal task number
  • forwarding the first data packet to the first computing node may include at least one of the following:
  • the first network function decomposes the computational task of the first data packet; for example, it can decompose the task according to a pre-assigned task decomposition method.
  • the first network function forwards the decomposed computing tasks to the corresponding first computing power node.
  • the communication method may further include:
  • the first network function receives the calculation result of the first data packet sent by the first computing node
  • the first network function sends the calculation result to the second network function and/or the terminal so that the second network function and/or the terminal can know the calculation result.
  • the first network function receives a notification message sent by the first computing node.
  • the notification message is used to notify that the calculation of the first data packet has been completed, so that the first network function can know that the calculation task has been completed.
  • the computing power information of computing power nodes can be reported through a registration process.
  • the communication method may further include at least one of the following:
  • the first network function receives a registration request sent by a computing power node; wherein the registration request is used to request the registration of the computing power resources of the computing power node, and/or, the registration request includes the computing power information of the computing power node;
  • the first network function stores the computing power information of computing nodes.
  • the computing power information of computing nodes can be proactively reported to the first network function (such as CCF, SMF, etc.) through the registration process.
  • the first network function such as CCF, SMF, etc.
  • the computing power information of the computing power node includes at least one of the following:
  • the types of computing nodes such as UPF, MEP, V2X server, 3rd APP, UE, etc.;
  • Identification of computing nodes such as IP address, FQDN, GPSI, etc.
  • Location information of computing nodes such as TAI or Geographic Area
  • computing resources of computing nodes such as APU, CPU, DPU, GPU, NPU, TPU, etc.;
  • the precision of computing resources of computing nodes such as MIPS, DMIPS, OPS, FLOPS, etc.
  • the available duration or failure time of computing resources of a computing node is the available duration or failure time of computing resources of a computing node.
  • the registration process can be triggered by a query
  • the communication method may further include:
  • the first network function sends an inquiry request to the computing power node, which inquires whether the computing power node needs to report its computing power resources. This allows the computing power node to promptly report its own computing power resources when required.
  • the process of receiving the registration request sent by the computing power node may include: the first network function receiving the registration request sent by the computing power node when it needs to report computing power resources.
  • the communication method may further include:
  • the first network function receives a deregistration request sent by a computing power node, the deregistration request being used to request to deregister the computing power resources of the computing power node.
  • the communication method may further include:
  • the first network function detects whether there are any ongoing computing services on the computing resources of the computing node, and performs any of the following:
  • the deregistration request may include at least one of the following:
  • the types of computing nodes such as UPF, MEP, V2X server, 3rd APP, UE, etc.;
  • Identification of computing nodes such as IP address, FQDN, GPSI, etc.
  • Location information of computing nodes such as TAI or Geographic Area
  • computing resources of computing nodes such as APU, CPU, DPU, GPU, NPU, TPU, etc.;
  • the precision of computing resources of computing nodes such as MIPS, DMIPS, OPS, FLOPS, etc.
  • the latest expiration time of computing resources of a computing node The latest expiration time of computing resources of a computing node.
  • Figure 2 illustrates the active registration process of computing resources in this embodiment of the present disclosure, and may include the following steps:
  • Step 22 CCF stores the computing power information of the computing power nodes locally.
  • FIG. 3 illustrates the paging registration process for computing resources in this embodiment of the present disclosure. This process may include the following steps:
  • Step 31 If a computing power control node, such as CCF, discovers (e.g., through other computing or resource distribution processes) that a computing power node in the network with computing power resources has not registered computing power with it, or that the validity period of previously reported computing power resources has expired, it can initiate an active query request to inquire whether the other party needs to (continue) report computing power resources.
  • a computing power control node such as CCF
  • a computing node determines that it has no computing resources to report, it will not initiate the registration process.
  • Step 34 CCF returns a registration response to the computing power node to notify that the computing power node's computing power resources have been successfully registered.
  • FIG. 4 illustrates the deregistration process of computing resources in this embodiment of the present disclosure.
  • This process may include the following steps:
  • Step 41 When the computing power on a computing power node (such as UPF, MEP, V2X server, 3rd APP, UE, etc.) of the mobile computing power network fails before the expected failure time, the computing power node actively sends a computing power deregistration request to the computing power control node such as CCF.
  • a computing power node such as UPF, MEP, V2X server, 3rd APP, UE, etc.
  • the request may carry at least one of the following information of the node: NF type (e.g., UPF, MEP, V2X server, 3rd APP, UE, etc.), NF ID (e.g., IP address, FQDN, GPSI, etc.), location information (e.g., TAI or Geographic Area, etc.), computing power resource type (e.g., APU, CPU, DPU, GPU, NPU, TPU, etc.), computing power resource precision (e.g., MIPS, DMIPS, OPS, FLOPS, etc.), computing power resource availability duration (or failure time), and latest failure time of computing power resource (much less than the original failure time).
  • NF type e.g., UPF, MEP, V2X server, 3rd APP, UE, etc.
  • NF ID e.g., IP address, FQDN, GPSI, etc.
  • location information e.g., TAI or Geographic Area, etc.
  • computing power resource type e.
  • Step 43 CCF returns a registration response to the computing power node to notify the computing power node that the computing power resources have been successfully registered.
  • Figure 5 is a flowchart of a communication method provided in an embodiment of this disclosure. The method is applied to a second network function. As shown in Figure 5, the method includes the following steps:
  • Step 51 The second network function sends the first request to the first network function.
  • the first requirement may be a computing power requirement or a computing requirement.
  • a computing power requirement can be expressed as a computing requirement; both have the same meaning.
  • the first network function can be selected from any of the following: Computing Control Function (CCF), Session Management Function (SMF), or Computing Network Element.
  • CCF Computing Control Function
  • SMF Session Management Function
  • NF Network Function
  • the first network function can also be a combined network function (NF) of CCF, SMF, and/or Computing Network Element, or other NFs capable of achieving similar functions; this embodiment does not limit this selection.
  • the second network function can be any of the following: User Plane Function (UPF) or Computing Node.
  • UPF User Plane Function
  • NF Computing Node
  • the second network function can also be a combined UPF and Computing Node (NF), or other NFs capable of achieving similar functionality; this embodiment does not limit this to any particular type.
  • the first requirement is the computational task requirement of the first service and/or the first data, or the first requirement is the computing power requirement of the first service and/or the first data.
  • the first service may be an immersive service such as AR, VR, or MR.
  • the first data may be immersive service data such as AR, VR, or MR.
  • the first requirement is used to match with registered computing resources.
  • Specific matching information and rules can be found in the above embodiments and will not be repeated here.
  • the first requirement may include, but is not limited to, at least one of the following:
  • the first task number is the task number within the network used to identify computational tasks.
  • Terminal location information which can be selected as TAI or Geographic Area, etc.
  • the computing type can be, but is not limited to, APU, CPU, DPU, GPU, NPU, TPU, etc.
  • the computational accuracy can be selected from, but is not limited to, MIPS, DMIPS, OPS, FLOPS, etc.
  • the solution in this embodiment can notify the first network function of the first requirement (such as computing power requirement or computing requirement), and then select a matching computing power node to calculate the business data, thereby reducing business lag, jitter and other issues, and improving the business experience.
  • the first network function of the first requirement such as computing power requirement or computing requirement
  • the communication method may further include at least one of the following:
  • the second network function receives a second data packet sent by the first device;
  • the first device is, for example, an application server (AS);
  • the second network function extracts the second requirement from the second data packet
  • the second network function maps the second task number, which is included in the second requirement, to the first task number; for example, when mapping the second task number to the first task number, the mapping can be random and there is no limitation on it.
  • the first task number is specifically an internal task number, that is, a task number used within the network to identify computational tasks, such as an identifier used within the 5GC.
  • the second task number is specifically an external task number, such as an identifier used between the AS, UE, and 5GC to identify tasks. Since external task numbers may be duplicated, the mapping between the second and first task numbers can prevent duplicate task numbers between different AFs. Furthermore, by using the mapping between the second and first task numbers, since the first task number is an internal network task number, external attacks can be prevented, thus improving security.
  • the second requirement may also include at least one of the following:
  • the computing type can be selected from, but is not limited to, APU, CPU, DPU, GPU, NPU, TPU, etc.;
  • the calculation accuracy can be selected from, but is not limited to, MIPS, DMIPS, OPS, FLOPS, etc.
  • the communication method may further include:
  • the second network function stores the aforementioned associations.
  • the terminal address, the address of the first device, and/or the terminal location can be associated with the corresponding task, thereby improving the accuracy of task execution.
  • the second network function receives a third data packet sent by the first device; the first device is, for example, an application server AS.
  • the second network function replaces the second task number carried in the third data packet with the corresponding first task number to obtain the first data packet;
  • the communication method may further include at least one of the following:
  • the second network function receives the calculation result of the first data packet sent by the first network function, so as to notify the terminal of the calculation result of the service data;
  • the second network function sends the calculation result of the first data packet to the terminal so that the terminal can know the calculation result of the business data.
  • Step 61 The first device sends a second data packet to the second network function; the second data packet carries a second requirement, which includes at least one of the following: second task number, calculation type, calculation accuracy, and estimated task duration.
  • the second network function can be any of the following: User Plane Function (UPF) or Computing Node.
  • the second network function can also be a combined UPF and Computing Node (NF), or other NFs capable of achieving similar functionality; this embodiment does not limit this to any particular type.
  • Step 2 After receiving this downlink data packet, UPF parses the packet header to find out that it contains computing power requirements. Then, it performs DPI and extracts the computing power requirements (i.e., the second requirements).
  • the computing power requirements may include at least one of the following: second task number (external task number), computing type (such as APU, CPU, DPU, GPU, NPU, TPU, etc.), computing precision (such as MIPS, DMIPS, OPS, FLOPS, etc.), and estimated task duration.
  • the UPF can map the second task number to the first task number (internal task number), such as using the superimposed information of the second task number + AF ID to map to the first task number (internal task number), and establish the association between the first/second task number and the UE IP address, AS IP address, and UE location, and record this association locally for use.
  • Step 3 The UPF forwards the computing power requirement (i.e., the first requirement) to the CCF via a data packet, which carries at least one of the following: first task number (internal task number), computing type (such as APU, CPU, DPU, GPU, NPU, TPU, etc.), computing precision (such as MIPS, DMIPS, OPS, FLOPS, etc.), and estimated task duration.
  • first task number internal task number
  • computing type such as APU, CPU, DPU, GPU, NPU, TPU, etc.
  • computing precision such as MIPS, DMIPS, OPS, FLOPS, etc.
  • the CCF sends a successful computing resource matching response to the UPF; if the CCF cannot find a suitable computing resource, it returns a computing resource matching failure message to the UPF.
  • the UPF receives a failure response, it sends a computing power request (i.e., the first request) to another CCF to request resource matching, and repeats steps 3-4.
  • a computing power request i.e., the first request
  • Step 5 AS sends a task/service data packet to UPF, the packet header of which carries the second task number (external task number) contained in the above computing power requirement (i.e., the second requirement).
  • Step 9a After receiving the calculation results, CCF forwards them to UPF.
  • Step 8b If the UE IP address is included when the computing power task is issued in step 7, the computing power node can forward the calculation result to the target UE based on the UE IP address after receiving the task data packet and performing the calculation.
  • Step 9b The compute node sends a notification message to the CCF to inform the CCF that the computation is complete.
  • Step 10b CCF notifies UPF that the calculation is complete.
  • the location of the CCF in the core network can be seen in Figure 8, but this embodiment is not limited thereto, and the CCF can be deployed based on actual needs.
  • the communication method provided in this disclosure can be executed by a communication device or a control module within that communication device for executing the communication method.
  • This disclosure uses the example of a communication device executing a communication method to illustrate the communication device provided in this disclosure.
  • Figure 9 is a schematic diagram of a communication device provided in an embodiment of this disclosure. This device is applied to a first network function and can be selected as CCF, SMF, computing power network element, etc.; as shown in Figure 9, the communication device 90 includes:
  • the first receiving module 91 is used to receive the first request sent by the second network function.
  • the communication device 90 also includes:
  • the matching module is used to match the first requirement with the registered computing resources to obtain the first computing node.
  • the first requirement is a computing power requirement or a computational requirement.
  • the communication device 90 further includes at least one of the following:
  • the second receiving module is used to receive the first data packet
  • the parsing module is used to parse the first task number carried in the first data packet
  • the first sending module is used to forward the first data packet to the first computing node according to the first task number.
  • the first sending module is specifically configured to perform at least one of the following:
  • the computational task of the first data packet is decomposed
  • the decomposed computing tasks are forwarded to the corresponding first computing power node.
  • the matching rules include at least one of the following:
  • the computing task is divided into multiple sub-time periods, and the tasks in the multiple sub-time periods are matched a second time.
  • the matching information includes at least one of the following:
  • NF Network Function
  • the first requirement includes at least one of the following:
  • First task number terminal location information, calculation type, calculation precision, and estimated task duration.
  • the communication device 90 further includes at least one of the following:
  • the third receiving module is used to receive a registration request sent by a computing power node; the registration request is used to request the registration of the computing power resources of the computing power node, and/or, the registration request includes the computing power information of the computing power node;
  • the storage module is used to store the computing power information of the computing nodes.
  • the computing power information of the computing power node includes at least one of the following:
  • the location information of the computing power node is the location information of the computing power node
  • the computing power resource type of the computing power node
  • the available duration or failure time of the computing power resources of the computing power node is the available duration or failure time of the computing power resources of the computing power node.
  • the communication device 90 also includes:
  • the second sending module is used to send an inquiry request to the computing power node, the inquiry request being used to ask the computing power node whether it needs to report computing power resources.
  • the third receiving module is specifically used to: receive the registration request sent by the computing power node when it needs to report computing power resources.
  • the communication device 90 also includes:
  • the fourth receiving module is used to receive a deregistration request sent by the computing power node, wherein the deregistration request is used to request to deregister the computing power resources of the computing power node.
  • the communication device 90 also includes:
  • the execution module is used to detect whether there are any ongoing computing services on the computing resources of the computing node, and to execute any of the following:
  • the computing power information of the computing power node is deleted after the service ends.
  • the service When there is a service being computed on the computing power resources of the computing power node, but the service cannot be terminated before the latest expiration time of the computing power resources, the service shall be migrated to another computing power node and the computing power information of the computing power node shall be deleted.
  • the deregistration request includes at least one of the following:
  • the location information of the computing power node is the location information of the computing power node
  • the computing power resource type of the computing power node
  • the available time or failure time of the computing power resources of the computing power node is the available time or failure time of the computing power resources of the computing power node
  • the latest expiration time of the computing resources of the computing node is the latest expiration time of the computing resources of the computing node.
  • the communication device 90 also includes:
  • the fifth receiving module is used to receive the calculation result of the first data packet sent by the first computing node
  • the third sending module is used to send the calculation result to the second network function and/or terminal.
  • the method further includes:
  • the sixth receiving module is used to receive notification information sent by the first computing node, the notification information being used to notify that the calculation of the first data packet is complete.
  • the first network function is any one of the following: Computing Power Control Function (CCF), Session Management Function (SMF), or Computing Power Network Element;
  • CCF Computing Power Control Function
  • SMF Session Management Function
  • Computing Power Network Element any one of the following:
  • the second network function is any one of the following: User Plane Function (UPF), or Computing Node.
  • UPF User Plane Function
  • Computing Node any one of the following: User Plane Function (UPF), or Computing Node.
  • the communication device 90 of this disclosure embodiment can implement the various processes of the method embodiment shown in FIG1 above and achieve the same technical effect. To avoid repetition, it will not be described again here.
  • Figure 10 is a schematic diagram of a communication device provided in an embodiment of this disclosure. This device is used for a second network function and can be selected as a UPF, computing node, etc. As shown in Figure 10, the communication device 100 includes:
  • the fourth sending module 101 is used to send the first request to the first network function.
  • the first requirement is the computing task requirement of the first service and/or the first data, or the first requirement is the computing power requirement of the first service and/or the first data.
  • the first requirement is used to match with registered computing resources.
  • the communication device 100 further includes at least one of the following:
  • the seventh receiving module is used to receive the second data packet sent by the first device
  • the extraction module is used to extract the second requirement from the second data packet
  • the mapping module is used to map the second task number included in the second requirement to the first task number.
  • the second task number is an external task number.
  • the mapping module is specifically used to: map the superimposed information of the second task number and the first identifier to the first task number; the first identifier is an application function identifier, which is used for the first device and/or terminal and/or the second network function.
  • the communication device 100 also includes:
  • a module is established to establish an association between the first task number and the second task number and at least one of the following: terminal address, address of the first device, and terminal location;
  • a storage module is used to store the aforementioned association relationships.
  • First task number terminal location information, calculation type, calculation precision, and estimated task duration.
  • the communication device 100 further includes at least one of the following:
  • the eighth receiving module is used to receive the third data packet sent by the first device
  • the replacement module is used to replace the second task number carried in the third data packet with the corresponding first task number to obtain the first data packet;
  • the fifth sending module is used to send the first data packet carrying the first task number to the first network function.
  • the communication device 100 further includes at least one of the following:
  • the ninth receiving module is used to receive the calculation result of the first data packet sent by the first network function
  • the sixth sending module is used to send the calculation result of the first data packet to the terminal.
  • the communication device 100 of this disclosure embodiment can implement the various processes of the method embodiment shown in FIG5 above and achieve the same technical effect. To avoid repetition, it will not be described again here.
  • Figure 11 is a schematic diagram of a communication device provided in an embodiment of this disclosure. This device is applied to a first device, such as an AS. As shown in Figure 11, the communication device 110 includes:
  • the seventh sending module 111 is used to send a second data packet to the second network function; wherein the second data packet carries a second requirement, the second requirement including at least one of the following: second task number, calculation type, calculation precision, and estimated task duration.
  • the communication device 110 of this disclosure embodiment can implement the various processes of the method embodiment shown in FIG6 above and achieve the same technical effect. To avoid repetition, it will not be described again here.
  • this embodiment of the present disclosure also provides a communication device 120, including a processor 121, a memory 122, and a program or instructions stored in the memory 122 and executable on the processor 121.
  • a communication device 120 including a processor 121, a memory 122, and a program or instructions stored in the memory 122 and executable on the processor 121.
  • the communication device 120 is a first network function
  • the program or instructions executed by the processor 121 implement the various processes of the communication method embodiment shown in FIG1 above, and achieve the same technical effect.
  • the communication device 120 is a second network function
  • the program or instructions executed by the processor 121 implement the various processes of the communication method embodiment shown in FIG5 above, and achieve the same technical effect.
  • the communication device 120 is a first device
  • the program or instructions executed by the processor 121 implement the various processes of the communication method embodiment shown in FIG6 above, and achieve the same technical effect. To avoid repetition, further details are omitted here.
  • This disclosure also provides a computer program product, including computer instructions.
  • the computer instructions When the computer instructions are executed by a processor, they can implement the various processes of the above-described communication method embodiments and achieve the same technical effects. To avoid repetition, they will not be described again here.
  • This disclosure also provides a readable storage medium storing a program or instructions.
  • the program or instructions When the program or instructions are executed by a processor, they can implement the various processes of the above-described communication method embodiments and achieve the same technical effects. To avoid repetition, further details are omitted here.
  • Computer-readable media include permanent and non-permanent, removable and non-removable media, which can store information by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc ROM (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device. As defined in this article, computer-readable media does not include transient media, such as modulated data signals and carrier waves.
  • PRAM phase-change memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • RAM random access memory
  • ROM read-only memory
  • EEPROM electrically

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本公开提供了一种通信方法、装置、通信设备及可读存储介质,属于通信技术领域。本公开实施例中的通信方法包括:第一网络功能接收第二网络功能发送的第一需求。

Description

通信方法、装置、通信设备及可读存储介质
相关申请的交叉引用
本公开主张在2024年05月17日在中国提交的中国专利申请号202410616124.6的优先权,其全部内容通过引用包含于此。
技术领域
本公开属于通信技术领域,具体涉及一种通信方法、装置、通信设备及可读存储介质。
背景技术
随着第三方应用的不断发展,增强现实(Augmented Reality,AR)、虚拟现实(Virtual Reality,VR)、混合现实(Mixed Reality,MR)等沉浸化业务市场快速增加,沉浸化业务需要网络对音频、视频、触觉、多种传感器信息等多模态数据进行传输与处理,并对传输速率、时延、可靠性等网络性能指标与处理器能力、存储能力等算力要求有严格的要求。但目前,受限于终端算力,沉浸化应用用户体验较差,无法完全实现用户的沉浸化体验,业务卡顿、抖动情况明显。这种情况下,如何提升业务体验是目前急需解决的问题。
发明内容
本公开实施例的目的是提供一种通信方法、装置、通信设备及可读存储介质,以解决如何提升业务体验的问题。
为了解决上述技术问题,本公开是这样实现的:
第一方面,提供了一种通信方法,应用于第一网络功能,包括:
第一网络功能接收第二网络功能发送的第一需求。
第二方面,提供了一种通信方法,应用于第二网络功能,包括:
第二网络功能向第一网络功能发送第一需求。
第三方面,提供了一种通信方法,应用于第一设备,包括:
第一设备向第二网络功能发送第二数据包;其中,所述第二数据包中携带第二需求,所述第二需求包括以下至少一项:第二任务号、计算类型、计算精度、预估任务时长。
第四方面,提供了一种通信装置,应用于第一网络功能,包括:
第一接收模块,用于接收第二网络功能发送的第一需求。
第五方面,提供了一种通信装置,应用于第二网络功能,包括:
第四发送模块,用于向第一网络功能发送第一需求。
第六方面,提供了一种通信装置,应用于第一设备,包括:
第七发送模块,用于向第二网络功能发送第二数据包;其中,所述第二数据包中携带第二需求,所述第二需求包括以下至少一项:第二任务号、计算类型、计算精度、预估任务时长。
第七方面,提供了一种通信设备,包括处理器、存储器及存储在所述存储器上并可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤,或者如第二方面所述的方法的步骤,或者如第三方面所述的方法的步骤。
第八方面,提供了一种可读存储介质,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如第一方面所述的方法的步骤,或者如第二方面所述的方法的步骤,或者如第三方面所述的方法的步骤。
第九方面,提供了一种计算机程序产品,包括计算机指令,所述计算机指令被处理器执行时实现如第一方面所述的方法的步骤,或者如第二方面所述的方法的步骤,或者如第三方面所述的方法的步骤。
通过本公开实施例中方案,可以实现将第一需求(如算力需求或计算需求)通知给第一网络功能,进而可以选出匹配的算力节点来计算业务数据,从而减少业务卡顿、抖动等情况,提升业务体验。
附图说明
图1是本公开实施例提供的一种通信方法的流程图;
图2是本公开实施例中算力资源的主动注册过程的流程图;
图3是本公开实施例中算力资源的寻呼注册过程的流程图;
图4是本公开实施例中算力资源的去注册过程的流程图;
图5是本公开实施例提供的另一种通信方法的流程图;
图6是本公开实施例提供的另一种通信方法的流程图;
图7是本公开实施例中算力需求的随路携带过程的流程图;
图8是本公开实施例中CCF的部署示意图;
图9是本公开实施例提供的一种通信装置的结构示意图;
图10是本公开实施例提供的一种通信装置的结构示意图;
图11是本公开实施例提供的一种通信装置的结构示意图;
图12是本公开实施例提供的一种通信设备的结构示意图。
具体实施方式
下面将结合本公开实施例中的附图,对本公开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公开一部分实施例,而不是全部的实施例。基于本公开中的实施例,本领域普通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,都属于本公开保护的范围。
本公开的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象,而不用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换,以便本公开的实施例能够以除了在这里图示或描述的那些以外的顺序实施,且“第一”、“第二”等所区分的对象通常为一类,并不限定对象的个数,例如第一对象可以是一个,也可以是多个。此外,说明书以及权利要求中“和/或”表示所连接对象的至少其中之一,字符“/”,一般表示前后关联对象是一种“或”的关系。
在本公开实施例中,算力节点可表述为计算节点,两者表示相同含义。
在本公开实施例中,算力需求可表述为计算需求,两者表示相同含义。
在本公开实施例中,算力任务可表述为计算任务,两者表示相同含义。
在本公开实施例中,算力资源可表述为计算资源,两者表示相同含义。
在本公开实施例中,算力资源可以包括移动算力资源。
下面结合附图,通过具体的实施例及其应用场景对本公开实施例提供的通信方法、装置、通信设备及可读存储介质进行详细地说明。
请参见图1,图1是本公开实施例提供的一种通信方法的流程图,该方法应用于第一网络功能,如图1所示,该方法包括如下步骤:
步骤11:第一网络功能接收第二网络功能发送的第一需求。
本公开实施例中,所述第一需求可以是算力需求或计算需求。算力需求可表述为计算需求,两者表示相同含义。
所述第一网络功能可选为以下任一项:算力控制功能(Computing Control Function,CCF)、会话管理功能(Session Management Function,SMF)、算力网元。此外,所述第一网络功能也可选为CCF、SMF和/或算力网元的合设网络功能(Network Function,NF),或者其他能够实现类似功能的NF,本实施例不对此进行限定。
所述第二网络功能可选为以下任一项:用户面功能(User Plane Function,UPF)、算力节点。此外,所述第二网络功能也可选为UPF和算力节点的合设NF,或者其他能够实现类似功能的NF,本实施例不对此进行限定。
可选的,所述第一需求可以包括但不限于以下至少一项:
第一任务号,所述第一任务号具体为内部任务号,即网络内部的用于识别计算任务的任务号,如用于第五代移动通信技术核心网(5th Generation Mobile Communication Technology Core,5GC)内部的标识;
终端位置信息,该终端位置信息可选为跟踪区识别码(Tracking Area identity,TAI)或地理区域(Geographic Area)等;
计算类型,该计算类型可选为但不限于加速处理单元(Accelerated Processing Unit,APU)、中央处理单元(Central Processing Unit,CPU)、数据处理单元(Data Processing Unit,DPU)、图形处理单元(Graphics Processing Unit,GPU)、神经网络处理单元(Neural Processing Unit,NPU)、张量处理单元(Tensor Processing Unit,TPU)等;
计算精度,该计算精度可选为但不限于每秒百万条指令数(Million Instructions Per Second,MIPS)、DMIPS(Dhrystone MIPS,表示CPU运行整数运算(Dhrystone)时所表现的相对性能)、每秒操作数(Operations Per Second,OPS)、每秒浮点运算次数(Floating Point Operations Per Second,FLOPS)等;
预估任务时长。
通过本公开实施例中方案,可以实现将第一需求(如算力需求或计算需求)通知给第一网络功能,进而可以选出匹配的算力节点来计算业务数据,从而减少业务卡顿、抖动等情况,提升业务体验。
可选的,所述通信方法还可以包括:
第一网络功能基于所述第一需求与已注册的算力资源进行匹配,获得第一算力节点。此第一算力节点可理解为匹配得到的用于执行所述第一需求中第一任务号对应的计算任务的算力节点,或者与所述第一需求中第一任务号对应的计算任务匹配的算力节点。由此可以选出合适的算力节点来执行算力任务。
可选的,所述匹配的信息可以包括以下至少一项:
a)算力资源类型,比如为APU、CPU、DPU、GPU、NPU、TPU等;
b)算力资源精度,比如为MIPS、DMIPS、OPS、FLOPS等;
c)网络功能标识NF ID,例如为IP地址(IP address)、完全限定域名(Fully Qualified Domain Name,FQDN)、全球产品和服务标识符(Global Product and Service Identifier,GPSI)等;
d)位置信息,比如包括算力节点的位置信息;
e)终端位置,比如为TAI或者Geographic Area等;
f)NF类型,例如为UPF、维护端点(Maintenance Endpoint,MEP)、车联网服务器(Vehicle-to-Everything server,V2X server)、第三方应用程序(Third-party Application,3rd APP)、用户设备(User Equipment,UE)等;
g)算力资源可用时长,此算力资源可用时长也可表述为失效时间。
也就是说,在匹配算力节点时,可以选择上述a)至g)中的部分或全部信息来进行匹配,以获得满足需求的算力节点。
可选的,所述匹配的规则可以包括但不限于以下至少一项:
(1)匹配算力资源类型与第一需求中的计算类型一致的算力节点,即若某算力节点的算力资源类型与第一需求中的计算类型一致,则该算力节点是匹配的;比如,此有关类型的条件是必须要满足的,即不能选择算力资源类型与计算类型不一致的算力节点;
(2)匹配算力资源精度与第一需求中的计算精度一致的算力节点,即若某算力节点的算力资源精度与第一需求中的计算精度一致,则该算力节点是匹配的;比如,此有关精度的条件是必须要满足的,即不能选择算力资源精度与计算精度不一致的算力节点;
(3)匹配算力资源与第二网络功能(如UPF)的距离在第一值范围内的算力节点;所述第一值可以由运营商根据实际需求预配置;
(4)优先匹配与第二网络功能的距离更近的算力节点;例如在同样满足计算类型和计算精度的情况下,优先匹配与第二网络功能(如UPF)的距离更近的算力节点;
(5)将与第二网络功能的距离大于第二值的算力节点视为不满足需求;所述第二值可以由运营商根据实际需求预配置,比如为N km等;例如,若某计算节点的算力资源与第二网络功能(如UPF)的距离大于Nkm,则将该计算节点视为不满足需求;
(6)当单个算力节点的可用时长不能满足第一需求中的预估任务时长时,将计算任务分成多个子时间段的任务,并对所述多个子时间段的任务进行二次匹配;比如若预估任务时长不能匹配,则可以将计算任务分成M个子时间段的任务进行二次匹配,M为大于1的整数。
需指出的,在匹配算力节点时,可以选择上述(1)至(6)中的部分或全部规则来匹配,本实施例不对此进行限定。
可选的,本公开实施例中的通信方法还可以包括以下至少一项:
所述第一网络功能接收第一数据包;
所述第一网络功能解析第一数据包中携带的第一任务号;
所述第一网络功能根据第一任务号,将第一数据包转发至第一算力节点。
比如,所述第一网络功能(如CCF)可以从UPF接收第一数据包;所述第一数据包可理解为业务数据包,其中携带有用于识别计算任务的第一任务号(即网络内部任务号)。
一种可选实施例中,所述第一网络功能(如CCF)在匹配得到的第一算力节点后,若后续接收到第一数据包,且该第一数据包中携带的第一任务号与该第一算力节点匹配,则可将该第一数据包转发至该第一算力节点,由第一算力节点执行计算任务。
可选的,上述将第一数据包转发至第一算力节点可以包括以下至少一项:
第一网络功能对所述第一数据包的计算任务进行分解;比如,可以按照已分配好的任务分解方式进行任务分解;
第一网络功能将分解后的计算任务转发至对应的第一算力节点。
可选的,在将所述第一数据包转发至第一算力节点之后,所述通信方法还可以还包括:
第一网络功能接收第一算力节点发送的对第一数据包的计算结果;
第一网络功能将所述计算结果发送给第二网络功能和/或终端,以便于第二网络功能和/或终端获知计算结果。
可选的,在将所述第一数据包转发至第一算力节点之后,所述通信方法还可以还包括:
第一网络功能接收第一算力节点发送的通知信息,所述通知信息用于通知完成对所述第一数据包的计算,以便于第一网络功能获知已完成计算任务。
本公开实施例中,可以通过注册流程来上报算力节点的算力信息。所述通信方法还可以包括以下至少一项:
第一网络功能接收算力节点发送的注册请求;其中,所述注册请求用于请求注册所述算力节点的算力资源,和/或,所述注册请求包含所述算力节点的算力信息;
第一网络功能存储算力节点的算力信息。
这样借助注册过程,可以主动地将算力节点的算力信息上报给第一网络功能(如CCF、SMF等)。
可选的,所述算力节点的算力信息包括以下至少一项:
算力节点的类型,比如为UPF、MEP、V2X server、3rd APP、UE等;
算力节点的标识,比如为IP地址、FQDN、GPSI等;
算力节点的位置信息,比如为TAI或者Geographic Area等;
算力节点的算力资源类型,比如为APU、CPU、DPU、GPU、NPU、TPU等;
算力节点的算力资源精度,比如为MIPS、DMIPS、OPS、FLOPS等;
算力节点的算力资源可用时长或失效时间。
可选的,可以借助问询来触发注册流程,所述通信方法还可以还包括:
第一网络功能向算力节点发送问询请求,所述问询请求用于询问所述算力节点是否需要上报算力资源。由此在算力节点需要上报算力资源时,可以及时上报自身的算力资源。
可选的,上述接收算力节点发送的注册请求的过程可以包括:第一网络功能接收算力节点在需要上报算力资源时发送的注册请求。
可选的,所述通信方法还可以还包括:
第一网络功能接收算力节点发送的去注册请求,所述去注册请求用于请求去注册所述算力节点的算力资源。
可选的,在接收去注册请求之后,所述通信方法还可包括:
第一网络功能检测所述算力节点的算力资源上是否有正在计算的业务,并执行以下任一项:
(A)当所述算力节点的算力资源上没有正在计算的业务时,删除所述算力节点的算力信息;
(B)当所述算力节点的算力资源上有正在计算的业务,且所述业务能够在所述算力资源的最晚失效时间之前结束时,在所述业务结束之后,删除所述算力节点的算力信息;
(C)当所述算力节点的算力资源上有正在计算的业务,但所述业务不能够在所述算力资源的最晚失效时间之前结束时,将所述业务迁移到其他算力节点,并删除所述算力节点的算力信息。
可选的,所述去注册请求可以包括以下至少一项:
算力节点的类型,比如为UPF、MEP、V2X server、3rd APP、UE等;
算力节点的标识,比如为IP地址、FQDN、GPSI等;
算力节点的位置信息,比如为TAI或者Geographic Area等;
算力节点的算力资源类型,比如为APU、CPU、DPU、GPU、NPU、TPU等;
算力节点的算力资源精度,比如为MIPS、DMIPS、OPS、FLOPS等;
算力节点的算力资源可用时长或失效时间;
算力节点的算力资源最晚失效时间。
下面结合图2至图4对本公开中的注册和去注册过程进行说明。
请参见图2,图2是本公开实施例中算力资源的主动注册过程,可以包括如下步骤:
步骤21:移动算力网络的算力节点(如UPF、MEP、V2X server、3rd APP、UE等)上有可用算力时,算力节点主动向CCF发送算力注册请求,其中可以携带本节点的以下至少一项:NF type(例如UPF、MEP、V2X server、3rd APP、UE等)、NF ID(例如IP address、FQDN、GPSI等)、位置信息(例如TAI或者Geographic Area等)、算力资源类型(例如APU、CPU、DPU、GPU、NPU、TPU等)、算力资源精度(例如MIPS、DMIPS、OPS、FLOPS等)、算力资源可用时长(或失效时间)。
步骤22:CCF将算力节点的算力信息存储在本地。
步骤23:CCF向算力节点返回注册响应,以通知算力节点的算力资源注册成功。
请参见图3,图3是本公开实施例中算力资源的寻呼注册过程,可以包括如下步骤:
步骤31:如果算力控制节点如CCF发现(例如通过其他计算或资源分发流程)有一处网络中的具有算力资源的算力节点并未向自己注册算力,或者是之前上报的算力资源的有效时间到期了,则可以发起主动问询请求(request),询问对方是否需要(继续)上报算力资源。
步骤32:若算力节点判断后有算力资源可以上报,则向CCF发送算力注册请求,其中可以携带本节点的以下至少一项:NF type(例如UPF、MEP、V2X server、3rd APP、UE等)、NF ID(例如IP address、FQDN、GPSI等)、位置信息(例如TAI或者Geographic Area等)、算力资源类型(例如APU、CPU、DPU、GPU、NPU、TPU等)、算力资源精度(例如MIPS、DMIPS、OPS、FLOPS等)、算力资源可用时长(或失效时间)。
而如果算力节点判断后没有算力资源可以上报,则不发起注册流程。
步骤33:CCF将算力节点的算力信息存储在本地。
步骤34:CCF向算力节点返回注册响应,以通知算力节点的算力资源注册成功。
请参见图4,图4是本公开实施例中算力资源的去注册过程,可以包括如下步骤:
步骤41:移动算力网络的算力节点(如UPF、MEP、V2X server、3rd APP、UE等)上的算力在预期失效时间前失效时,算力节点主动向算力控制节点如CCF发送算力去注册请求,其中可以携带本节点的以下至少一项:NF type(例如UPF、MEP、V2X server、3rd APP、UE等)、NF ID(例如IP address、FQDN、GPSI等)、位置信息(例如TAI或者Geographic Area等)、算力资源类型(例如APU、CPU、DPU、GPU、NPU、TPU等)、算力资源精度(例如MIPS、DMIPS、OPS、FLOPS等)、算力资源可用时长(或失效时间)、算力资源最晚失效时间(远小于原定失效时间)。
步骤42:CCF在收到去注册请求后,检查算力资源上是否有正在计算的业务,相关业务是否在最晚失效时间之前可以结束;如果可以,等待业务全部结束,删除本地算力资源信息;如果不可以,将算力业务迁移到其他合适的算力节点上,全部算力业务结束后,删除本地算力资源信息。
步骤43.CCF向算力节点返回去注册响应,以通知算力节点的算力资源去注册成功。
请参见图5,图5是本公开实施例提供的一种通信方法的流程图,该方法应用于第二网络功能,如图5所示,该方法包括如下步骤:
步骤51:第二网络功能向第一网络功能发送第一需求。
本公开实施例中,所述第一需求可以是算力需求或计算需求。算力需求可表述为计算需求,两者表示相同含义。
所述第一网络功能可选为以下任一项:算力控制功能CCF、会话管理功能SMF、算力网元。此外,所述第一网络功能也可选为CCF、SMF和/或算力网元的合设网络功能NF,或者其他能够实现类似功能的NF,本实施例不对此进行限定。
所述第二网络功能可选为以下任一项:用户面功能UPF、算力节点。此外,所述第二网络功能也可选为UPF和算力节点的合设NF,或者其他能够实现类似功能的NF,本实施例不对此进行限定。
可选的,所述第一需求是第一业务和/或第一数据对计算任务的需求,或者,所述第一需求是第一业务和/或第一数据对算力任务的需求。所述第一业务可选为AR、VR、MR等沉浸化业务。所述第一数据可选为AR、VR、MR等沉浸化业务数据。
可选的,所述第一需求用于与已注册的算力资源进行匹配。具体匹配的信息和规则可以参见上述实施例中所述,在此不再赘述。
可选的,所述第一需求可以包括但不限于以下至少一项:
第一任务号,所述第一任务号具体为内部任务号,即网络内部的用于识别计算任务的任务号;
终端位置信息,该终端位置信息可选为TAI或Geographic Area等;
计算类型,该计算类型可选为但不限于APU、CPU、DPU、GPU、NPU、TPU等;
计算精度,该计算精度可选为但不限于MIPS、DMIPS、OPS、FLOPS等;
预估任务时长。
通过本公开实施例中方案,可以实现将第一需求(如算力需求或计算需求)通知给第一网络功能,进而可以选出匹配的算力节点来计算业务数据,从而减少业务卡顿、抖动等情况,提升业务体验。
可选的,所述通信方法还可以包括以下至少一项:
第二网络功能接收第一设备发送的第二数据包;所述第一设备比如为应用服务器(Application Server,AS)等;
第二网络功能从第二数据包中提取第二需求;
第二网络功能将第二需求包括的第二任务号映射为第一任务号;比如,在将第二任务号映射为第一任务号时,可以随机映射,对此不作限定。
需指出的,所述第一任务号具体为内部任务号,即网络内部的用于识别计算任务的任务号,如用于5GC内部的标识。所述第二任务号具体为外部任务号,如用于AS、UE与5GC之间用于识别任务的标识号。由于外部任务号可能存在重复的情况,因此借助第二任务号与第一任务号的映射,可以防止不同AF之间的任务号重复的情况。此外借助第二任务号与第一任务号的映射,由于第一任务号为网络内部任务号,故可以防止外部攻击,提高安全性。
可选的,所述第二需求除了包括第二任务号之外,还可包括以下至少一项:
计算类型,可选为但不限于APU、CPU、DPU、GPU、NPU、TPU等;
计算精度,可选为但不限于MIPS、DMIPS、OPS、FLOPS等;
预估任务时长。
可选的,上述将第二需求包括的第二任务号映射为第一任务号的方式可以包括:第二网络功能将所述第二任务号与第一标识的叠加信息映射为第一任务号;所述第一标识为应用功能标识,所述应用功能标识用于第一设备和/或终端和/或第二网络功能。由于外部任务号(即第二任务号)可能存在重复的情况,而应用功能标识通常为唯一标识,因此将第二任务号与第一标识的叠加信息映射为第一任务号,可以防止混淆情况的出现。
可选的,所述通信方法还可以包括:
第二网络功能建立第一任务号和第二任务号与以下至少一项之间的关联关系:终端地址、第一设备的地址、终端位置;
第二网络功能存储所述关联关系。
这样,借助上述建立/存储的关联关系,可以将终端地址、第一设备的地址和/或终端位置与相应任务关联,从而提高任务执行的准确性。
可选的,所述通信方法还可以包括以下至少一项:
第二网络功能接收第一设备发送的第三数据包;所述第一设备比如为应用服务器AS等;
第二网络功能将第三数据包中携带的第二任务号替换为对应的第一任务号,得到第一数据包;
第二网络功能将携带第一任务号的第一数据包发送给第一网络功能,以便第一网络功能根据该第一任务号,将第一数据包转发至对应的算力节点来执行计算任务。
可选的,所述通信方法还可以包括以下至少一项:
第二网络功能接收第一网络功能发送的对第一数据包的计算结果,以便于通知终端对业务数据的计算结果;
第二网络功能将对第一数据包的计算结果发送给终端,以便于终端获知对业务数据的计算结果。
请参见图6,图6是本公开实施例提供的一种通信方法的流程图,该方法应用于第一设备,所述第一设备比如为AS等。如图6所示,该方法包括如下步骤:
步骤61:第一设备向第二网络功能发送第二数据包;所述第二数据包中携带第二需求,所述第二需求包括以下至少一项:第二任务号、计算类型、计算精度、预估任务时长。
本公开实施例中,所述第二网络功能可选为以下任一项:用户面功能UPF、算力节点。此外,所述第二网络功能也可选为UPF和算力节点的合设NF,或者其他能够实现类似功能的NF,本实施例不对此进行限定。
所述第二需求是第一业务和/或第一数据对计算任务的需求,或者,所述第二需求是第一业务和/或第一数据对算力任务的需求。所述第一业务可选为AR、VR、MR等沉浸化业务。所述第一数据可选为AR、VR、MR等沉浸化业务数据。
可选的,所述第二任务号具体为外部任务号,如用于AS、UE与5GC之间用于识别任务的标识号。所述计算类型可选为但不限于APU、CPU、DPU、GPU、NPU、TPU等;所述计算精度可选为但不限于MIPS、DMIPS、OPS、FLOPS等。
通过本公开实施例中方案,可以由第一设备将相关的计算/算力需求随数据包下发至网络,从而使得网络选出匹配的算力节点来计算业务数据,从而减少业务卡顿、抖动等情况,提升业务体验。
请参见图7,本公开具体实施例中算力需求的随路携带过程可以包括:
步骤1:UE与第三方服务器AS之间开始进行XR等计算需求较大的业务,AS在大量计算包开始的前一个或者前N个数据包的包头中随路携带算力需求(即第二需求),并下发该前一个或者前N个数据包至UPF;该包头进行标记(flag),说明包含算力需求(比如包含第二任务号、计算类型、计算精度、预估任务时长)。
步骤2:UPF在收到这个下行数据包后,解析其中的包头,获知其包含算力需求,则进行DPI并提取出算力需求(即第二需求),其中,该算力需求可包含以下至少一项:第二任务号(外部任务号)、计算类型(如APU、CPU、DPU、GPU、NPU、TPU等)、计算精度(如MIPS、DMIPS、OPS、FLOPS等)、预估任务时长。
可选的,UPF可将第二任务号映射为第一任务号(内部任务号),如使用第二任务号+AF ID的叠加信息映射为第一任务号(内部任务号),并建立第一/第二任务号与UE IP地址、AS IP地址、UE位置之间的关联关系,并将该关联关系记录在本地,以便使用。
步骤3:UPF通过数据包将算力需求(即第一需求)转发给CCF,其中携带以下至少一项:第一任务号(内部任务号)、计算类型(如APU、CPU、DPU、GPU、NPU、TPU等)、计算精度(如MIPS、DMIPS、OPS、FLOPS等)、预估任务时长。
步骤4:CCF在接收到算力需求(即第一需求)后,基于本地已经注册的算力资源与UPF上报的算力需求,进行匹配。具体匹配规则可以参见上述实施例,在此不再赘述。
可选的,CCF在匹配好后,向UPF发送算力资源匹配成功响应;而如果CCF无法匹配到合适的计算资源,则向UPF返回算力资源匹配失败消息。
可选的,UPF如果收到失败响应,则向另一个CCF发送算力需求(即第一需求),请求资源匹配,重复步骤3-4。
步骤5:AS向UPF发送任务/业务数据包,其包头携带上述算力需求(即第二需求)中包含的第二任务号(外部任务号)。
步骤6:UPF在接收到任务数据包后,解析发现数据包头中携带约定好的第二任务号,则将该第二任务号替换为对应的第一任务号(内部任务号),并将携带第一任务号(内部任务号)的数据包转发给CCF。
步骤7:CCF收到数据包后,解析包头中携带的第一任务号(内部任务号),并将数据包转发至对应的步骤4匹配得到的算力节点,即执行计算任务分发。
在上述步骤7之后,可存在以下两种情况:
情况A:
步骤8a:算力节点在收到数据包后进行计算,并将计算后的大数据包即计算结果反馈给CCF。
步骤9a:CCF收到计算结果后转发给UPF。
步骤10a:UPF收到计算结果后转发给UE。
情况B:
步骤8b:若步骤7下发算力任务时携带UE IP地址,则算力节点收到任务数据包并计算后,可根据UE IP地址将计算结果转发给目标UE。
步骤9b:计算节点向CCF发送通知信息,用于通知CCF计算完成。
步骤10b:CCF通知UPF计算完成。
一种可选实施例中,CCF在核心网中的位置可以参见图8所示,但本实施例不以此为限,可以基于实际需求部署CCF。
需要说明的是,本公开实施例提供的通信方法,执行主体可以为通信装置,或者该通信装置中的用于执行通信方法的控制模块。本公开实施例中以通信装置执行通信方法为例,说明本公开实施例提供的通信装置。
请参见图9,图9是本公开实施例提供的一种通信装置的结构示意图,该装置应用于第一网络功能,可选为CCF、SMF、算力网元等;如图9所示,通信装置90包括:
第一接收模块91,用于接收第二网络功能发送的第一需求。
可选的,通信装置90还包括:
匹配模块,用于基于所述第一需求与已注册的算力资源进行匹配,获得第一算力节点。
可选的,所述第一需求是算力需求或计算需求。
可选的,通信装置90还包括以下至少一项:
第二接收模块,用于接收第一数据包;
解析模块,用于解析第一数据包中携带的第一任务号;
第一发送模块,用于根据第一任务号,将第一数据包转发至第一算力节点。
可选的,所述第一发送模块具体用于执行以下至少一项:
对所述第一数据包的计算任务进行分解;
将分解后的计算任务转发至对应的第一算力节点。
可选的,所述匹配的规则包括以下至少一项:
匹配算力资源类型与所述第一需求中的计算类型一致的算力节点;
匹配算力资源精度与所述第一需求中的计算精度一致的算力节点;
匹配算力资源与所述第二网络功能的距离在第一值范围内的算力节点;
优先匹配与所述第二网络功能的距离更近的算力节点;
将与所述第二网络功能的距离大于第二值的算力节点视为不满足需求;
当单个算力节点的可用时长不能满足所述第一需求中的预估任务时长时,将计算任务分成多个子时间段的任务,并对所述多个子时间段的任务进行二次匹配。
可选的,所述匹配的信息包括以下至少一项:
算力资源类型;
算力资源精度;
网络功能标识NF ID;
位置信息;
终端位置;
网络功能NF类型;
算力资源可用时长。
可选的,所述第一需求包括以下至少一项:
第一任务号、终端位置信息、计算类型、计算精度、预估任务时长。
可选的,通信装置90还包括以下至少一项:
第三接收模块,用于接收算力节点发送的注册请求;所述注册请求用于请求注册所述算力节点的算力资源,和/或,所述注册请求包含所述算力节点的算力信息;
存储模块,用于存储算力节点的算力信息。
可选的,所述算力节点的算力信息包括以下至少一项:
所述算力节点的类型;
所述算力节点的标识;
所述算力节点的位置信息;
所述算力节点的算力资源类型;
所述算力节点的算力资源精度;
所述算力节点的算力资源可用时长或失效时间。
可选的,通信装置90还包括:
第二发送模块,用于向所述算力节点发送问询请求,所述问询请求用于询问所述算力节点是否需要上报算力资源。
可选的,所述第三接收模块具体用于:接收所述算力节点在需要上报算力资源时发送的所述注册请求。
可选的,通信装置90还包括:
第四接收模块,用于接收算力节点发送的去注册请求,所述去注册请求用于请求去注册所述算力节点的算力资源。
可选的,通信装置90还包括:
执行模块,用于检测所述算力节点的算力资源上是否有正在计算的业务,并执行以下任一项:
当所述算力节点的算力资源上没有正在计算的业务时,删除所述算力节点的算力信息;
当所述算力节点的算力资源上有正在计算的业务,且所述业务能够在所述算力资源的最晚失效时间之前结束时,在所述业务结束之后,删除所述算力节点的算力信息;
当所述算力节点的算力资源上有正在计算的业务,但所述业务不能够在所述算力资源的最晚失效时间之前结束时,将所述业务迁移到其他算力节点,并删除所述算力节点的算力信息。
可选的,所述去注册请求包括以下至少一项:
所述算力节点的类型;
所述算力节点的标识;
所述算力节点的位置信息;
所述算力节点的算力资源类型;
所述算力节点的算力资源精度;
所述算力节点的算力资源可用时长或失效时间;
所述算力节点的算力资源最晚失效时间。
可选的,通信装置90还包括:
第五接收模块,用于接收所述第一算力节点发送的对所述第一数据包的计算结果;
第三发送模块,用于将所述计算结果发送给所述第二网络功能和/或终端。
可选的,所述方法还包括:
第六接收模块,用于接收所述第一算力节点发送的通知信息,所述通知信息用于通知完成对所述第一数据包的计算。
可选的,所述第一网络功能为以下任一项:算力控制功能CCF、会话管理功能SMF、算力网元;
和/或,所述第二网络功能为以下任一项:用户面功能UPF、算力节点。
本公开实施例的通信装置90,可以实现上述图1所示的方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
请参见图10,图10是本公开实施例提供的一种通信装置的结构示意图,该装置应用于第二网络功能,可选为UPF、算力节点等;如图10所示,通信装置100包括:
第四发送模块101,用于向第一网络功能发送第一需求。
可选的,所述第一需求是第一业务和/或第一数据对计算任务的需求,或者,所述第一需求是第一业务和/或第一数据对算力任务的需求;
和/或,所述第一需求用于与已注册的算力资源进行匹配。
可选的,通信装置100还包括以下至少一项:
第七接收模块,用于接收第一设备发送的第二数据包;
提取模块,用于从第二数据包中提取第二需求;
映射模块,用于将第二需求包括的第二任务号映射为第一任务号。
可选的,所述第一任务号是内部任务号;
和/或,所述第二任务号是外部任务号。
可选的,所述映射模块具体用于:将所述第二任务号与第一标识的叠加信息映射为所述第一任务号;所述第一标识为应用功能标识,所述应用功能标识用于第一设备和/或终端和/或所述第二网络功能。
可选的,通信装置100还包括:
建立模块,用于建立所述第一任务号和所述第二任务号与以下至少一项之间的关联关系:终端地址、所述第一设备的地址、终端位置;
存储模块,用于存储所述关联关系。
可选的,所述第二需求还包括以下至少一项:
计算类型、计算精度、预估任务时长。
可选的,所述第一需求包括以下至少一项:
第一任务号、终端位置信息、计算类型、计算精度、预估任务时长。
可选的,通信装置100还包括以下至少一项:
第八接收模块,用于接收第一设备发送的第三数据包;
替换模块,用于将第三数据包中携带的第二任务号替换为对应的第一任务号,得到第一数据包;
第五发送模块,用于将携带第一任务号的第一数据包发送给所述第一网络功能。
可选的,通信装置100还包括以下至少一项:
第九接收模块,用于接收第一网络功能发送的对第一数据包的计算结果;
第六发送模块,用于将对第一数据包的计算结果发送给终端。
本公开实施例的通信装置100,可以实现上述图5所示的方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
请参见图11,图11是本公开实施例提供的一种通信装置的结构示意图,该装置应用于第一设备,比如AS等,如图11所示,通信装置110包括:
第七发送模块111,用于向第二网络功能发送第二数据包;其中,所述第二数据包中携带第二需求,所述第二需求包括以下至少一项:第二任务号、计算类型、计算精度、预估任务时长。
本公开实施例的通信装置110,可以实现上述图6所示的方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
可选的,如图12所示,本公开实施例还提供一种通信设备120,包括处理器121,存储器122,存储在存储器122上并可在所述处理器121上运行的程序或指令,例如,该通信设备120为第一网络功能时,该程序或指令被处理器121执行时实现上述图1所示的通信方法实施例的各个过程,且能达到相同的技术效果。或者,该通信设备120为第二网络功能时,该程序或指令被处理器121执行时实现上述图5所示的通信方法实施例的各个过程,且能达到相同的技术效果。或者,该通信设备120为第一设备时,该程序或指令被处理器121执行时实现上述图6所示的通信方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
本公开实施例还提供了一种计算机程序产品,包括计算机指令,所述计算机指令被处理器执行时可实现上述通信方法实施例的各个过程且能达到相同的技术效果,为避免重复,这里不再赘述。
本公开实施例还提供了一种可读存储介质,其上存储有程序或指令,所述程序或指令被处理器执行时可实现上述通信方法实施例的各个过程且能达到相同的技术效果,为避免重复,这里不再赘述。
计算机可读介质包括永久性和非永久性、可移动和非可移动媒体,可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机的存储介质的例子包括,但不限于相变内存(Programmable RAM,PRAM)、静态随机存取存储器(Static RAM,SRAM)、动态随机存取存储器(Dynamic RAM,DRAM)、其他类型的随机存取存储器(Random Access Memory,RAM)、只读存储器(Read-Only Memory,ROM)、电可擦除可编程只读存储器(Electrically EPROM,EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(Compact Disc ROM,CD-ROM)、数字多功能光盘(Digital Versatile Disc,DVD)或其他光学存储、磁盒式磁带,磁带磁盘存储或其他磁性存储设备或任何其他非传输介质,可用于存储可以被计算设备访问的信息。按照本文中的界定,计算机可读介质不包括暂存电脑可读媒体(transitory media),如调制的数据信号和载波。
需要说明的是,在本文中,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。
上述本公开实施例序号仅仅为了描述,不代表实施例的优劣。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件,但很多情况下前者是更佳的实施方式。基于这样的理解,本公开的技术方案本质上或者说对相关技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中,包括若干指令用以使得一台服务分类设备(可以是手机,计算机,服务器,空调器,或者网络设备等)执行本公开各个实施例所述的方法。
以上所述仅是本公开的优选实施方式,应当指出,对于本技术领域的普通技术人员来说,在不脱离本公开原理的前提下,还可以做出若干改进和润饰,这些改进和润饰也应视为本公开的保护范围。

Claims (35)

  1. 一种通信方法,包括:
    第一网络功能接收第二网络功能发送的第一需求。
  2. 根据权利要求1所述的方法,所述方法还包括:
    所述第一网络功能基于所述第一需求与已注册的算力资源进行匹配,获得第一算力节点。
  3. 根据权利要求1或2所述的方法,其中,所述第一需求是算力需求或计算需求。
  4. 根据权利要求1或2所述的方法,所述方法还包括以下至少一项:
    所述第一网络功能接收第一数据包;
    所述第一网络功能解析第一数据包中携带的第一任务号;
    所述第一网络功能根据第一任务号,将第一数据包转发至第一算力节点。
  5. 根据权利要求4所述的方法,其中,所述将第一数据包转发至第一算力节点,包括以下至少一项:
    所述第一网络功能对所述第一数据包的计算任务进行分解;
    所述第一网络功能将分解后的计算任务转发至对应的第一算力节点。
  6. 根据权利要求2所述的方法,其中,所述匹配的规则包括以下至少一项:
    匹配算力资源类型与所述第一需求中的计算类型一致的算力节点;
    匹配算力资源精度与所述第一需求中的计算精度一致的算力节点;
    匹配算力资源与所述第二网络功能的距离在第一值范围内的算力节点;
    优先匹配与所述第二网络功能的距离更近的算力节点;
    将与所述第二网络功能的距离大于第二值的算力节点视为不满足需求;
    当单个算力节点的可用时长不能满足所述第一需求中的预估任务时长时,将计算任务分成多个子时间段的任务,并对所述多个子时间段的任务进行二次匹配。
  7. 根据权利要求2所述的方法,其中,所述匹配的信息包括以下至少一项:
    算力资源类型;
    算力资源精度;
    网络功能标识NF ID;
    位置信息;
    终端位置;
    网络功能NF类型;
    算力资源可用时长。
  8. 根据权利要求1所述的方法,其中,所述第一需求包括以下至少一项:
    第一任务号、终端位置信息、计算类型、计算精度、预估任务时长。
  9. 根据权利要求1所述的方法,所述方法还包括以下至少一项:
    所述第一网络功能接收算力节点发送的注册请求;所述注册请求用于请求注册所述算力节点的算力资源,和/或,所述注册请求包含所述算力节点的算力信息;
    所述第一网络功能存储算力节点的算力信息。
  10. 根据权利要求9所述的方法,其中,所述算力节点的算力信息包括以下至少一项:
    所述算力节点的类型;
    所述算力节点的标识;
    所述算力节点的位置信息;
    所述算力节点的算力资源类型;
    所述算力节点的算力资源精度;
    所述算力节点的算力资源可用时长或失效时间。
  11. 根据权利要求9所述的方法,所述方法还包括:
    所述第一网络功能向所述算力节点发送问询请求,所述问询请求用于询问所述算力节点是否需要上报算力资源。
  12. 根据权利要求11所述的方法,其中,所述第一网络功能接收算力节点发送的注册请求,包括:
    所述第一网络功能接收所述算力节点在需要上报算力资源时发送的所述注册请求。
  13. 根据权利要求1所述的方法,所述方法还包括:
    所述第一网络功能接收算力节点发送的去注册请求,所述去注册请求用于请求去注册所述算力节点的算力资源。
  14. 根据权利要求13所述的方法,所述方法还包括:
    所述第一网络功能检测所述算力节点的算力资源上是否有正在计算的业务,并执行以下任一项:
    当所述算力节点的算力资源上没有正在计算的业务时,删除所述算力节点的算力信息;
    当所述算力节点的算力资源上有正在计算的业务,且所述业务能够在所述算力资源的最晚失效时间之前结束时,在所述业务结束之后,删除所述算力节点的算力信息;
    当所述算力节点的算力资源上有正在计算的业务,但所述业务不能够在所述算力资源的最晚失效时间之前结束时,将所述业务迁移到其他算力节点,并删除所述算力节点的算力信息。
  15. 根据权利要求13所述的方法,其中,所述去注册请求包括以下至少一项:
    所述算力节点的类型;
    所述算力节点的标识;
    所述算力节点的位置信息;
    所述算力节点的算力资源类型;
    所述算力节点的算力资源精度;
    所述算力节点的算力资源可用时长或失效时间;
    所述算力节点的算力资源最晚失效时间。
  16. 根据权利要求4所述的方法,所述方法还包括:
    所述第一网络功能接收所述第一算力节点发送的对所述第一数据包的计算结果;
    所述第一网络功能将所述计算结果发送给所述第二网络功能和/或终端。
  17. 根据权利要求4所述的方法,所述方法还包括:
    所述第一网络功能接收所述第一算力节点发送的通知信息,所述通知信息用于通知完成对所述第一数据包的计算。
  18. 根据权利要求1所述的方法,其中,所述第一网络功能为以下任一项:算力控制功能CCF、会话管理功能SMF、算力网元;
    和/或,
    所述第二网络功能为以下任一项:用户面功能UPF、算力节点。
  19. 一种通信方法,包括:
    第二网络功能向第一网络功能发送第一需求。
  20. 根据权利要求19所述的方法,其中,所述第一需求是第一业务和/或第一数据对计算任务的需求,或者,所述第一需求是第一业务和/或第一数据对算力任务的需求;
    和/或,所述第一需求用于与已注册的算力资源进行匹配。
  21. 根据权利要求19或20所述的方法,所述方法还包括以下至少一项:
    所述第二网络功能接收第一设备发送的第二数据包;
    所述第二网络功能从第二数据包中提取第二需求;
    所述第二网络功能将第二需求包括的第二任务号映射为第一任务号。
  22. 根据权利要求21所述的方法,其中,
    所述第一任务号是内部任务号;
    和/或,
    所述第二任务号是外部任务号。
  23. 根据权利要求21所述的方法,其中,所述第二网络功能将第二需求包括的第二任务号映射为第一任务号,包括:
    所述第二网络功能将所述第二任务号与第一标识的叠加信息映射为所述第一任务号;所述第一标识为应用功能标识,所述应用功能标识用于第一设备和/或终端和/或所述第二网络功能。
  24. 根据权利要求21所述的方法,所述方法还包括:
    所述第二网络功能建立所述第一任务号和所述第二任务号与以下至少一项之间的关联关系:终端地址、所述第一设备的地址、终端位置;
    所述第二网络功能存储所述关联关系。
  25. 根据权利要求21所述的方法,其中,所述第二需求还包括以下至少一项:
    计算类型、计算精度、预估任务时长。
  26. 根据权利要求19所述的方法,其中,所述第一需求包括以下至少一项:
    第一任务号、终端位置信息、计算类型、计算精度、预估任务时长。
  27. 根据权利要求19所述的方法,所述方法还包括以下至少一项:
    所述第二网络功能接收第一设备发送的第三数据包;
    所述第二网络功能将第三数据包中携带的第二任务号替换为对应的第一任务号,得到第一数据包;
    所述第二网络功能将携带第一任务号的第一数据包发送给所述第一网络功能。
  28. 根据权利要求19所述的方法,所述方法还包括以下至少一项:
    所述第二网络功能接收所述第一网络功能发送的对第一数据包的计算结果;
    所述第二网络功能将对第一数据包的计算结果发送给终端。
  29. 一种通信方法,包括:
    第一设备向第二网络功能发送第二数据包;其中,所述第二数据包中携带第二需求,所述第二需求包括以下至少一项:第二任务号、计算类型、计算精度、预估任务时长。
  30. 一种通信装置,包括:
    第一接收模块,用于接收第二网络功能发送的第一需求。
  31. 一种通信装置,包括:
    第四发送模块,用于向第一网络功能发送第一需求。
  32. 一种通信装置,包括:
    第七发送模块,用于向第二网络功能发送第二数据包;其中,所述第二数据包中携带第二需求,所述第二需求包括以下至少一项:第二任务号、计算类型、计算精度、预估任务时长。
  33. 一种通信设备,包括处理器,存储器及存储在所述存储器上并可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求1至18任一项所述的方法的步骤,或者如权利要求19至28任一项所述的方法的步骤,或者如权利要求29所述的方法的步骤。
  34. 一种可读存储介质,其中,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如权利要求1至18任一项所述的方法的步骤,或者如权利要求19至28任一项所述的方法的步骤,或者如权利要求29所述的方法的步骤。
  35. 一种计算机程序产品,包括计算机指令,所述计算机指令被处理器执行时实现如权利要求1至18任一项所述的方法的步骤,或者如权利要求19至28任一项所述的方法的步骤,或者如权利要求29所述的方法的步骤。
PCT/CN2025/094224 2024-05-17 2025-05-12 通信方法、装置、通信设备及可读存储介质 Pending WO2025237231A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202410616124.6A CN120980073A (zh) 2024-05-17 2024-05-17 通信方法、装置、通信设备及可读存储介质
CN202410616124.6 2024-05-17

Publications (1)

Publication Number Publication Date
WO2025237231A1 true WO2025237231A1 (zh) 2025-11-20

Family

ID=97651244

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2025/094224 Pending WO2025237231A1 (zh) 2024-05-17 2025-05-12 通信方法、装置、通信设备及可读存储介质

Country Status (2)

Country Link
CN (1) CN120980073A (zh)
WO (1) WO2025237231A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200076884A1 (en) * 2018-08-31 2020-03-05 Nakamoto & Turing Labs Inc Methods and apparatus for performing distributed computing using blockchain
CN114828039A (zh) * 2021-01-21 2022-07-29 维沃移动通信有限公司 数据传输方法、终端及网络侧设备
CN115277598A (zh) * 2022-06-20 2022-11-01 中国联合网络通信集团有限公司 一种算力资源调度方法、装置及计算机可读存储介质
CN116761216A (zh) * 2023-07-07 2023-09-15 中国电信股份有限公司技术创新中心 算力路由确定方法、装置和移动核心网络
CN117119477A (zh) * 2022-05-16 2023-11-24 中国移动通信有限公司研究院 数据处理方法、装置、相关设备及存储介质
CN117320081A (zh) * 2022-06-21 2023-12-29 维沃移动通信有限公司 算力处理方法、装置及通信设备

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200076884A1 (en) * 2018-08-31 2020-03-05 Nakamoto & Turing Labs Inc Methods and apparatus for performing distributed computing using blockchain
CN114828039A (zh) * 2021-01-21 2022-07-29 维沃移动通信有限公司 数据传输方法、终端及网络侧设备
CN117119477A (zh) * 2022-05-16 2023-11-24 中国移动通信有限公司研究院 数据处理方法、装置、相关设备及存储介质
CN115277598A (zh) * 2022-06-20 2022-11-01 中国联合网络通信集团有限公司 一种算力资源调度方法、装置及计算机可读存储介质
CN117320081A (zh) * 2022-06-21 2023-12-29 维沃移动通信有限公司 算力处理方法、装置及通信设备
CN116761216A (zh) * 2023-07-07 2023-09-15 中国电信股份有限公司技术创新中心 算力路由确定方法、装置和移动核心网络

Also Published As

Publication number Publication date
CN120980073A (zh) 2025-11-18

Similar Documents

Publication Publication Date Title
CN116405461B (zh) 一种数据处理方法、网元设备以及可读存储介质
US10897431B2 (en) Cloud resource processing method and physical node
CN109842906B (zh) 一种通信的方法、装置及系统
CN110580256B (zh) 一种识别应用标识的方法、设备及系统
US8903972B2 (en) Method and apparatus for sharing contents using information of group change in content oriented network environment
CN116668511B (zh) 一种数据处理方法、网元设备以及可读存储介质
CN114071649B (zh) 访问本地网络的方法和装置
WO2023274293A1 (zh) 算力通告的发送方法、装置及算力网元节点
CN115349119A (zh) 用于在网络中部署网络功能(nf)集时的增强的5gc恢复的方法和装置
WO2021056314A1 (zh) 一种为分布式服务提供请求响应确切通信时延保障的系统
CN112218342A (zh) 一种实现核心网子切片容灾的方法、装置和系统
CN107070790B (zh) 一种路由学习方法及路由设备
CN106899554A (zh) 一种防止arp欺骗的方法及装置
WO2025237231A1 (zh) 通信方法、装置、通信设备及可读存储介质
US8498401B2 (en) Mobile-to-mobile call determination
CN111385371B (zh) Mac地址获取方法、装置和设备
US12587588B2 (en) Information processing method and apparatus and device
WO2021164236A1 (zh) 报文的处理方法及装置
CN116095180B (zh) 一种日志回传的选路方法、装置和存储介质
CN102857352B (zh) 基于叠加网的组播、广播方法及系统
CN106506176B (zh) 一种策略和计费规则生成方法和系统
CN116761159A (zh) 设备多归属接入方法、装置、计算机设备和存储介质
CN118101592A (zh) 网络地址结构、卫星及其报文处理方法、装置
CN119520588B (zh) 信令处理方法、装置、系统、计算机设备和可读存储介质
CN109089000B (zh) 语音呼叫处理方法及装置

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: 25802879

Country of ref document: EP

Kind code of ref document: A1