WO2024041472A1 - 一种通信方法、装置、设备和存储介质 - Google Patents

一种通信方法、装置、设备和存储介质 Download PDF

Info

Publication number
WO2024041472A1
WO2024041472A1 PCT/CN2023/113956 CN2023113956W WO2024041472A1 WO 2024041472 A1 WO2024041472 A1 WO 2024041472A1 CN 2023113956 W CN2023113956 W CN 2023113956W WO 2024041472 A1 WO2024041472 A1 WO 2024041472A1
Authority
WO
WIPO (PCT)
Prior art keywords
cpu
information
cloud platform
network function
cpu core
Prior art date
Application number
PCT/CN2023/113956
Other languages
English (en)
French (fr)
Inventor
李婷
张晓华
倪蔚辰
孙奇
李男
刘建华
池刚毅
Original Assignee
中国移动通信有限公司研究院
中国移动通信集团有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国移动通信有限公司研究院, 中国移动通信集团有限公司 filed Critical 中国移动通信有限公司研究院
Publication of WO2024041472A1 publication Critical patent/WO2024041472A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network

Definitions

  • the present disclosure relates to the field of wireless communications, and in particular, to a communication method, device, equipment and storage medium.
  • operating systems such as X86 and Advanced RISC Machines (ARM) have a central processing unit (CPU) frequency adjustment function, which can expand the maximum frequency or minimum frequency through the system configuration CPU, thereby Adjust the CPU frequency based on CPU utilization, CPU load and other conditions.
  • CPU central processing unit
  • a cloud virtualization platform is defined in the Open Radio Access Network (O-RAN) architecture, which is used to deploy wireless network functions using general physical resources through cloudification or virtualization technology, including O -Cloud Infrastructure Management Services (IMS), O-Cloud Deployment Management Services (DMS), Virtualized Network Function (VNF) and/or Containerized Network Function (Cloud Native Network) Functions, CNF).
  • O-Cloud can interact with Service Management and Orchestration (SMO) through the O2 interface, so that SMO manages resources such as O-Cloud infrastructure, VNF and/or CNF.
  • SMO uses the O2ims interface for infrastructure and cloud platform deployment and resource management
  • SMO uses the O2dms interface for VNF and/or CNF deployment and resource management.
  • the main purpose of the present disclosure is to provide a communication method, apparatus, equipment and storage medium.
  • Embodiments of the present disclosure provide a communication method, which is applied to network elements.
  • the method includes:
  • the first information is used to indicate the computing resource configuration of the network function of the cloud platform, and the first information is determined based on at least measurement information of the cloud platform.
  • the first information includes at least one of the following:
  • the second information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the second information includes at least one of the following:
  • the method also includes:
  • the measurement information includes at least one of the following:
  • the third information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the third information includes at least one of the following:
  • the network function includes at least one of the following:
  • Near-RT RIC Near real-time wireless intelligent controller
  • O-CU Open Radio Access Network Control Unit
  • O-CU-CP Open Radio Access Network Control Unit Control Plane
  • O-CU-UP Open Radio Access Network Control Unit User Plane
  • O-DU Open Radio Access Network Distributed Unit
  • Open Radio Access Network Radio Unit (O-RU).
  • determining the first information based on the measurement information includes at least one of the following:
  • the first rule includes at least one of the following:
  • the measurement information includes a second trigger condition corresponding to each parameter value combination and a second operation corresponding to the second trigger condition.
  • the first trigger condition includes: the parameter value belongs to the first threshold interval;
  • the second trigger condition includes: each parameter value in the parameter value combination belongs to the second threshold interval corresponding to the parameter value.
  • the first operation and the second operation include at least one of the following:
  • the network element includes at least one of the following: service management and orchestration (Service Management and Orchestration (SMO), Network Functions Virtualization Orchestrator (NFVO), Mobile Edge Orchestrator (ME orchestrator, MEO), Virtualized Network Function Manager (VNFM), Container Network Function Management Container Network Function Manager (CNFM).
  • SMO Service Management and Orchestration
  • NFVO Network Functions Virtualization Orchestrator
  • ME orchestrator Mobile Edge Orchestrator
  • VNFM Virtualized Network Function Manager
  • CFM Container Network Function Management Container Network Function Manager
  • Embodiments of the present disclosure provide a communication method applied to a cloud platform.
  • the method includes:
  • Receive first information from a network element the first information is used to indicate the computing resource configuration of the network function of the cloud platform, and the first information is determined based on at least measurement information of the cloud platform.
  • the first information includes at least one of the following:
  • the second information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the second information includes at least one of the following:
  • the method also includes:
  • the measurement information includes at least one of the following:
  • the third information is at least used to indicate the binding relationship between the virtual resources of the cloud platform and the CPU core;
  • the third information includes at least one of the following:
  • the network function includes at least one of the following:
  • the method also includes:
  • Configuration of computing resources of the cloud platform network function is performed according to the first information.
  • configuring the computing resources of the cloud platform network function based on the first information includes at least one of the following:
  • the network element includes at least one of the following: SMO, NFVO, MEO, VNFM, and CNFM.
  • An embodiment of the present disclosure provides a communication device, which is applied to a network element.
  • the device includes: a first sending module, configured to send first information to a cloud platform; the first information is used to indicate the calculation of network functions of the cloud platform. Resource configuration, the first information is determined based on at least the measurement information of the cloud platform.
  • the first information includes at least one of the following:
  • the second information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the second information includes at least one of the following:
  • the device further includes:
  • a first receiving module configured to receive measurement information from the cloud platform
  • a first determination module configured to determine the first information based on the measurement information
  • the measurement information includes at least one of the following:
  • the third information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the third information includes at least one of the following:
  • the network function includes at least one of the following:
  • the first determination module is used to perform at least one of the following:
  • the first rule includes at least one of the following:
  • the measurement information includes a second trigger condition corresponding to each parameter value combination and a second operation corresponding to the second trigger condition.
  • the first trigger condition includes: the parameter value belongs to the first threshold interval;
  • the second trigger condition includes: each parameter value in the parameter value combination belongs to the second threshold interval corresponding to the parameter value.
  • the first operation and the second operation include at least one of the following:
  • the network element includes at least one of the following: SMO, NFVO, MEO, VNFM, and CNFM.
  • An embodiment of the present disclosure provides a communication device applied to a cloud platform.
  • the device includes: a second receiving module for receiving first information from a network element; the first information is used to indicate the cloud platform network
  • the computing resource configuration of the function, the first information is determined based on at least the measurement information of the cloud platform.
  • the first information includes at least one of the following:
  • the second information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the second information includes at least one of the following:
  • the device further includes: a second sending module, configured to send measurement information to the network element;
  • the measurement information includes at least one of the following:
  • the third information is at least used to indicate the binding relationship between the virtual resources of the cloud platform and the CPU core;
  • the third information includes at least one of the following:
  • the network function includes at least one of the following:
  • the device further includes: a configuration module configured to configure computing resources of the cloud platform network function according to the first information.
  • the configuration module is used to perform at least one of the following:
  • the network element includes at least one of the following: SMO, NFVO, MEO, VNFM, and CNFM.
  • An embodiment of the present disclosure also provides a communication device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any The steps of the method described in the above item; or, when the processor executes the program, the steps of any of the methods described in the cloud platform side are implemented.
  • Embodiments of the present disclosure also provide a computer-readable storage medium on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of any one of the methods on the network element side are implemented; or, When the computer program is executed by the processor, the steps of any one of the methods on the cloud platform side are implemented.
  • a communication method, device, equipment and storage medium provided by embodiments of the present disclosure.
  • the method includes: a network element sends first information to a cloud platform; the first information is used to indicate the computing resource configuration of the network function of the cloud platform. , the first information is determined based on at least the measurement information of the cloud platform; the cloud platform receives the first information from the network element.
  • the network element can determine the first information based on the cloud platform's own measurement information, and the cloud platform flexibly configures the computing resources of the network function based on the first information, realizing the adjustment of computing resources for the cloud platform network function level to achieve more accurate energy saving. configuration.
  • Figure 1 is a schematic diagram of an O-RAN architecture
  • Figure 2 is a schematic flow chart of a communication method provided by an embodiment of the present disclosure
  • Figure 3 is a schematic flow chart of another communication method provided by an embodiment of the present disclosure.
  • Figure 4 is a schematic flowchart of a communication method provided by an application embodiment of the present disclosure.
  • Figure 5 is a schematic flowchart of a method for generating an energy-saving policy provided by an application embodiment of the present disclosure
  • Figure 6 is a schematic flowchart of another method for generating an energy-saving policy provided by an application embodiment of the present disclosure
  • Figure 7 is a flow chart of a CPU frequency configuration policy generation method provided by an application embodiment of the present disclosure. Process diagram
  • FIG. 8 is a schematic flowchart of another CPU frequency configuration policy generation method provided by an application embodiment of the present disclosure.
  • Figure 9 is a schematic flowchart of a method for generating a CPU core configuration policy provided by an application embodiment of the present disclosure
  • Figure 10 is a schematic flowchart of a method for generating a CPU power state configuration policy provided by an application embodiment of the present disclosure
  • Figure 11 is a schematic structural diagram of a communication device provided by an embodiment of the present disclosure.
  • Figure 12 is a schematic structural diagram of another communication device provided by an embodiment of the present disclosure.
  • Figure 13 is a schematic structural diagram of a communication device provided by an embodiment of the present disclosure.
  • O-RAN architecture is shown in Figure 1.
  • O-Cloud can interact with SMO through the O2 interface.
  • SMO When deploying VNF and/or CNF, SMO configures the number of CPU cores in the configuration file sent to O-Cloud DMS through O2dms, thereby triggering O-Cloud DMS to create and instantiate the resources required for VNF and/or CNF.
  • the configuration file sent by SMO to O-Cloud DMS through O2dms only specifies the number of virtual CPUs (vCPU).
  • vCPU virtual CPUs
  • VNF and/or CNF The resources allocated by VNF and/or CNF, such as reducing the CPU frequency, modifying the CPU power status, etc., cannot realize the adjustment of O-Cloud's CPU frequency, CPU status and other resources, nor can it be flexibly based on O-Cloud's resource occupancy. configuration and/or energy-saving operation.
  • embodiments of the present disclosure provide a communication method.
  • the network element sends first information to the cloud platform; the first information is used to indicate the computing resource configuration of the cloud platform network function, and the first information is based on at least the cloud platform.
  • the measurement information of the platform is determined; the cloud platform receives the first information from the network element. This enables the cloud platform to flexibly configure computing resources for network functions based on its own measurement information.
  • Figure 2 is a schematic flowchart of a communication method provided by an embodiment of the present disclosure; as shown in Figure 2, the method can be applied to network elements, and the method includes:
  • Step 201 Send first information to the cloud platform; the first information is used to indicate the computing resource configuration of the network function of the cloud platform, and the first information is determined based at least on the measurement information of the cloud platform.
  • the network element includes but is not limited to at least one of the following: Service Management and Orchestration (SMO), Network Functions Virtualization Orchestrator (NFVO), Mobile Edge Orchestrator (ME orchestrator, MEO), Virtualized Network Function Manager (VNFM), Container Network Function Manager (CNFM).
  • SMO Service Management and Orchestration
  • NFVO Network Functions Virtualization Orchestrator
  • ME orchestrator Mobile Edge Orchestrator
  • MEO Mobile Edge Orchestrator
  • VNFM Virtualized Network Function Manager
  • CNFM Container Network Function Manager
  • the network element can also be a network device or network function, which can interact with the cloud platform.
  • it can be an SMO, or a certain functional unit in the SMO.
  • the computing resource configuration at least includes: CPU configuration.
  • the first information includes but is not limited to at least one of the following:
  • the second information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the second information includes, but is not limited to, at least one of the following:
  • the network element instructs the cloud platform to configure network function computing resources through first information.
  • the first information may include relevant parameters required for operation, such as at least one of the following: second information, the number of virtual CPUs, CPU frequency options or CPU frequency mode, CPU power status.
  • the cloud platform performs relevant resource configuration based on the parameters included in the received first information, thereby achieving effects such as energy saving.
  • the cloud platform can be implemented through one or more CPUs.
  • the CPU is a physical CPU, and the physical CPU has one or more CPU cores (also called CPU cores), that is, the CPU can be a single-core CPU, a multi-core CPU, etc., and the corresponding CPU can have a single CPU core or multiple CPU cores.
  • CPU cores also called CPU cores
  • One or more virtual CPUs can be implemented on the CPU.
  • virtualization technology When applied, virtualization technology is used to allocate corresponding virtual CPUs to network functions on the CPU core bound to the network function based on the mapping relationship between virtual CPUs and CPU cores.
  • the number of virtual CPUs may be the number of virtual CPUs allocated by the CPU for network functions. If the cloud platform implements multiple network functions, the number of virtual CPUs may include: the number of virtual CPUs allocated for each network function.
  • CU Centralized Unit
  • DU Distributed Unit
  • RAN Intelligent Controller Radio Access Network Intelligent Controller
  • RIC Radio Access Network Intelligent Controller
  • Virtualization technology implements corresponding software functions, and virtualized wireless network functions can be deployed on general servers, which are called network functions here.
  • Network functions can be allocated specific computing, storage, network and other resources on the physical server, and the network functions and allocated resources together constitute an instance.
  • the network functions include at least one of the following:
  • Near-RT RIC Near real-time wireless intelligent controller
  • O-RAN Central Unit O-CU
  • O-CU Control Plane O-CU-CP
  • O-CU User Plane O-CU User Plane, O-CU-UP
  • O-RAN Distributed Unit Open Radio Access Network Distributed Unit
  • Open Radio Access Network Radio Unit O-RAN Radio Unit, O-RU
  • O-RAN Radio Unit O-RU
  • O-RAN refers to Open Radio Access Network.
  • the frequency options of the CPU or the frequency mode of the CPU may include but are not limited to one of the following:
  • scaling set speed scaling_set_speed
  • schedutil the kernel space scheduler calculates the load of each task, and the CPU frequency controller adjusts the CPU frequency based on CPU utilization
  • the power state of the CPU can include but is not limited to one of the following:
  • G-State global system-level power state, which corresponds to other states
  • P-State system performance setting state, reducing CPU energy consumption by adjusting the CPU frequency
  • P0 is the highest performance
  • P1, P2, etc. are states of saving power
  • T-State energy saving state, energy saving is achieved by adjusting CPU running time
  • the binding rules of the CPU core can include but are not limited to one of the following: STATIC, DYNAMIC.
  • STATIC can refer to static; DYNAMIC can refer to dynamic.
  • CPU frequency option or CPU frequency mode, CPU power status, CPU core binding rules, etc. can be preset or configured based on actual application scenarios, and the above description does not limit them.
  • the cloud platform can inform the network element of its own status, and the network element provides an energy-saving strategy and/or a computing resource adjustment strategy based on the status of the cloud platform, that is, sending the first information.
  • the method further includes:
  • the measurement information includes but is not limited to at least one of the following:
  • the third information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the cloud platform measures its own status to obtain measurement information; reports the measurement information to the network element, and the network element determines the first information based on the measurement information.
  • the CPU utilization may refer to the utilization of the CPU core and/or the utilization of the physical CPU; the CPU frequency may refer to the frequency of the CPU core and/or the frequency of the physical CPU; here No restrictions.
  • the temperature of the CPU generally refers to the temperature of the physical CPU; the power state of the CPU generally refers to the power state of the physical CPU.
  • the third information includes, but is not limited to, at least one of the following:
  • the cloud platform can monitor its own CPU frequency, CPU utilization, CPU temperature, CPU power status, and the status of the currently running network functions (such as whether the network function supports CPU core binding, the CPU core to which the network function is bound, The corresponding relationship between virtual CPU and CPU core), the number of virtual CPUs (which can specifically include the number of virtual CPUs allocated for each network function), etc. are detected to obtain the measurement information, which can be used to generate energy saving information for the network element after being notified to the network element.
  • Strategy the status of the currently running network functions (such as whether the network function supports CPU core binding, the CPU core to which the network function is bound, The corresponding relationship between virtual CPU and CPU core), the number of virtual CPUs (which can specifically include the number of virtual CPUs allocated for each network function), etc.
  • the cloud platform obtains measurement information after detection, and informs the network element of its current status through the measurement information; the network element determines the first information based on the measurement information, and sends the first information to the cloud platform; the cloud platform performs network operations based on the first information.
  • Configuration of functional computing resources For example, bind a certain network function to the CPU core, adjust the CPU frequency option or CPU frequency mode to high-performance mode, etc.
  • the priority of resource adjustment policy parameters involved in the energy-saving policy can be set. For example, when generating an energy-saving policy, first adjust the CPU frequency or adjust the number of CPU cores used, and finally modify the power status of the CPU.
  • the measurement information and/or configuration information may also include cluster-related identification information.
  • the cluster-related configuration information includes but is not limited to at least one of the following:
  • Cloud ID which is the identifier of the cloud platform, is used to identify the cloud platform where the current network function is located;
  • the first identifier is used to identify the master node of the cloud platform cluster where the current network function is located;
  • the second identifier is used to identify the working node of the cloud platform cluster where the current network function is located;
  • the third identifier is used to identify the instance of the current network function.
  • determining the first information based on the measurement information includes:
  • the preset first rule and the preset first model can be formulated based on any requirements, and there are no limitations on the formulation method and content of the first rule and/or the first model.
  • the first rule includes but is not limited to at least one of the following:
  • the measurement information includes a second trigger condition corresponding to each parameter value combination and a second operation corresponding to the second trigger condition.
  • the first trigger condition includes: the parameter value belongs to the first threshold interval;
  • the second trigger condition includes: each parameter value in the parameter value combination belongs to the second threshold interval corresponding to the parameter value.
  • the first operation and the second operation include, but are not limited to, at least one of the following:
  • the parameter values included in the measurement information may be real-time measurement values or statistical values of any parameter in the measurement information
  • the parameter value combination may include real-time measurement values or statistical values of multiple parameters in the measurement information.
  • the parameter value may include the value of at least one of the following parameters: CPU frequency, CPU utilization, and CPU temperature.
  • Different first trigger conditions can be set for each parameter, and different second trigger conditions can also be set for each parameter combination. Specifically, different threshold intervals can be set for each parameter.
  • setting the corresponding threshold interval for the CPU frequency may include: one or a group of threshold intervals for CPU utilization, such as [0, f1), [f1, f2), ... [fn-1, fn];
  • threshold interval which may include: one or a group of threshold intervals for CPU utilization, such as [0, t1), [t1, t2), ... [tn-1, tn];
  • the corresponding threshold interval which may include: one or a group of CPU temperature threshold intervals, such as [0, p1), [p1, p2), ... [pn-1, pn].
  • the resource occupancy of the CPU bound to the network function such as the threshold interval to which the CPU frequency belongs, CPU utilization The threshold interval to which the rate belongs, the threshold interval to which the CPU temperature belongs
  • a certain network function of a certain cloud platform is allocated to CPU core 1 and CPU core 2; based on the measurement information, it is determined that the CPU utilization of CPU core 1 belongs to the threshold interval [t1, t2) and the CPU utilization of CPU core 2. rate belongs to the threshold interval [0, t1) (t2 is greater than t1, t1 is the lowest threshold of utilization), then it can be determined that the operation is to adjust the CPU core bound to the network function, for example: unbinding the relationship between CPU core 2 and the network function , release CPU core 2.
  • a certain network function of a certain cloud platform is allocated CPU core 1; it is determined based on the measurement information that the CPU utilization of CPU core 1 belongs to the threshold interval [t2, t3), and the CPU frequency belongs to the threshold interval [f1, f2 ); you can determine that the operation is to adjust the CPU frequency option or the CPU frequency mode, for example: adjust from performance (high-performance mode) to powersave (energy-saving mode), or adjust to custom mode (userspace mode).
  • threshold intervals parameters value combinations, and executed operations are only examples, and the trigger conditions, threshold intervals, and corresponding first and second operations are not limited.
  • the above operation notifies the cloud platform through the first information to realize the configuration of the computing resources of the cloud platform.
  • the method further includes:
  • the training sample set includes multiple training samples, and each training sample includes sample measurement information and a label; the label may be the first information of the sample.
  • the sample measurement information can be set with reference to the above-mentioned measurement information, and the sample first information can be set with reference to the above-mentioned first information, which will not be described in detail here.
  • receiving measurement information from the cloud platform includes at least one of the following:
  • the measurement information includes at least one of the following:
  • the at least one parameter may include: CPU frequency, CPU utilization, CPU temperature, CPU power status, etc.
  • satisfying the preset conditions includes:
  • the measurement value and/or statistical value of at least one parameter in the measurement information exceeds a preset threshold.
  • the statistical value can be the maximum value, minimum value, average value, total value, etc. within a time period, and can be set based on actual needs.
  • the first information includes at least one of the following:
  • the first information when the first information includes a performance mode, the first information is used to indicate that the cloud platform network function computing resources are increased or kept unchanged to support high-performance operation; or,
  • the first information is used to indicate that the cloud platform network function computing resources are increased or decreased within a preset range to support adaptive adjustment; or,
  • the first information When the first information includes a power management mode or an energy saving mode, the first information is used to indicate that the cloud platform network function computing resources are reduced to support energy saving.
  • the computing resource adjustment parameters used for the cloud platform network function computing resource adjustment include at least one of the following:
  • the cloud platform network function computing resource adjustment may include at least one of the following: The cloud platform network function computing resources increase or remain unchanged to achieve high-performance operation; the cloud platform network function computing resources increase or decrease within a preset range to achieve adaptive resource adjustment; the cloud platform network function computing resources decrease , achieve energy saving.
  • embodiments of the present disclosure also provide a communication method on the cloud platform side.
  • Figure 3 is a schematic flow chart of another communication method provided by an embodiment of the present disclosure; as shown in Figure 3, the method can be applied to the cloud platform, and the method includes:
  • Step 301 Receive first information from a network element; the first information is used to indicate the computing resource configuration of the network function of the cloud platform, and the first information is determined based at least on the measurement information of the cloud platform.
  • the network element includes at least one of the following: SMO, NFVO, MEO, VNFM, and CNFM.
  • the cloud platform may be O-Cloud, VIM, CIM and other cloud platforms.
  • the first information includes but is not limited to at least one of the following:
  • the second information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the second information includes, but is not limited to, at least one of the following:
  • the method further includes:
  • the measurement information includes but is not limited to at least one of the following:
  • the third information is at least used to indicate the binding relationship between the virtual resources of the cloud platform and the CPU core;
  • the third information includes, but is not limited to, at least one of the following:
  • the network functions include but are not limited to at least one of the following:
  • the cloud platform after receiving the first information, configures the computing resources of the network function according to the instructions of the first information.
  • the method further includes:
  • Configure the computing resources of the cloud platform network function according to the first information may include but is not limited to at least one of the following:
  • the above adjustment may be specifically performed based on the content included in the first information.
  • the cloud platform may determine and indicate at least one of the above operations based on the first information.
  • sending measurement information to the network element includes at least one of the following:
  • measurement information is sent to the network element.
  • the measurement information includes at least one of the following:
  • satisfying the preset conditions includes:
  • the measurement value and/or statistical value of at least one parameter in the measurement information exceeds a preset threshold.
  • the network element implements network function level adjustments for one or several CPU cores (such as CPU frequency options or CPU frequency modes, CPU Compared with the method of adjusting the CPU frequency of all CPU cores through BIOS for physical servers, it has a more fine-grained and customized effect and reduces unnecessary CPU energy consumption. overhead, thereby achieving the purpose of energy saving on the cloud platform.
  • CPU cores such as CPU frequency options or CPU frequency modes, CPU
  • Embodiments of the present disclosure also provide an O-RAN architecture, which includes: a cloud platform and network elements; the cloud platform may be O-Cloud, and the network element may be SMO.
  • O-RAN architecture which includes: a cloud platform and network elements; the cloud platform may be O-Cloud, and the network element may be SMO.
  • SMO can combine network function load conditions to generate and deliver energy-saving policies to O-Cloud DMS, allowing O-Cloud DMS to dynamically adjust the configuration of the CPU bound to network functions and reduce O-Cloud DMS.
  • -Cloud energy consumption to achieve the purpose of energy saving.
  • Figure 4 is a schematic flow chart of an energy saving method provided by an application embodiment of the present disclosure; as shown in Figure 4, the method includes:
  • Step 401 O-Cloud IMS reports the current first measurement information to SMO through the O2ims interface (401.1) and O-Cloud DMS through the O2dms interface (401.2).
  • the first measurement information may include at least one of the following:
  • the third information is used to indicate the binding relationship between VNF and/or CNF and the currently bound CPU core
  • Resource usage of the CPU core bound to the VNF and/or CNF (including CPU frequency, CPU load, CPU utilization, CPU temperature, CPU power status, etc.).
  • the third information may include at least one of the following: whether the VNF and/or CNF supports the CPU Core binding; at least one CPU core bound to VNF and/or CNF; the corresponding relationship between virtual CPU and CPU core.
  • the network function is VNF and/or CNF as an example.
  • the network function can also include Near-RT RIC, O-CU, O-CU-CP, O-CU-UP, O-DU, O -RU et al.
  • Reporting methods include: periodic reporting and/or event-triggered reporting.
  • Step 402 SMO generates an energy-saving policy (at least including the energy-saving policy of O-Cloud) based on the first measurement information reported by O-Cloud IMS and O-Cloud DMS;
  • SMO can also combine the measurement information reported by the VNF and/or CNF through the O1 interface (which can include network service load, such as the number of users, the number of PRBs, PRB utilization, etc.) to generate an energy-saving policy.
  • O1 interface which can include network service load, such as the number of users, the number of PRBs, PRB utilization, etc.
  • Step 403. Based on the energy-saving policy, SMO delivers the first configuration information to O-Cloud DMS through the O2dms interface;
  • the first configuration information is used to instruct O-Cloud to perform energy-saving configuration, including but not limited to at least one of the following: CPU frequency mode or CPU frequency option, second information, number of virtual CPUs, CPU power status and other parameters;
  • the second information includes at least one of the following: whether the VNF and/or CNF supports CPU core binding; at least one CPU core bound by the VNF and/or CNF; the binding rule of the CPU core; and the corresponding relationship between the virtual CPU and the CPU core.
  • Step 404 O-Cloud DMS updates the deployment of VNF and/or CNF according to the first configuration information sent by SMO;
  • a deployment management request may be sent, and the deployment management request carries the first configuration information.
  • Step 405 O-Cloud DMS sends a success or failure response to SMO after updating.
  • the first measurement information reported through the O2ims interface and/or the O2dms interface and the first configuration information issued through the O2dms interface may include cluster-related identification information, as shown in Table 1.
  • masterNodeId refers to the master node ID
  • workerNodeId refers to the worker node ID
  • instanceId refers to the instance ID.
  • O-Cloud IMS reports the first measurement information to SMO through the O2ims interface and O-Cloud DMS through the O2dms interface to inform the current resource occupancy of VNF and/or CNF.
  • the reported first measurement information includes but Not limited to the parameters shown in Table 2:
  • the configuration parameters in the first configuration information delivered by SMO to O-Cloud DMS through the O2dms interface include but are not limited to those shown in Table 3:
  • Bios refers to the Basic Input Output System.
  • the CPU frequency mode or CPU frequency option is adjusted during the operation of the operating system, which can include but is not limited to one of the following: performance (high-performance mode, executed according to the configured maximum frequency); powersave (energy-saving mode, executed according to the configured minimum frequency) Execution); userspace (executed according to the CPU frequency set by user space, need to carry the target CPU frequency configuration parameters, for example, scaling_set_speed); schedutil (kernel space scheduler calculates the load of each task, and the CPU frequency controller adjusts the CPU based on CPU utilization frequency); ondemand (adjust the CPU frequency in proportion according to the CPU load); conservative (adjust the CPU frequency in steps according to the CPU load).
  • the power state of the CPU can include but is not limited to one of the following: G-State (global system-level power state, which corresponds to other states); S-State (CPU sleep state, the larger the number, the higher the degree of sleep); C -State (valid only for G0 state, CPU core level working state, the larger the number, the higher the degree of energy saving); P-State (system performance setting state, reducing CPU energy consumption by adjusting the CPU frequency, P0 is the highest performance, P1, P2 and other states are power-saving states); T-State (energy-saving state, energy saving is achieved by adjusting the CPU running time).
  • G-State global system-level power state, which corresponds to other states
  • S-State CPU sleep state, the larger the number, the higher the degree of sleep
  • C -State valid only for G0 state, CPU core level working state, the larger the number, the higher the degree of energy saving
  • P-State system performance setting state, reducing CPU energy consumption by adjusting the CPU frequency
  • P0
  • CPU core binding strategies can include: STATIC, DYNAMIC, etc.
  • the cloud platform energy-saving method proposed in this disclosed embodiment enhances the O2dms interface to deliver configuration parameters such as CPU frequency to O-Cloud DMS, so that O-Cloud DMS can implement energy-saving strategies, thereby solving the problem of SMO through the O2dms interface.
  • the problem that the configuration file delivered to O-Cloud DMS only specifies the number of vCPUs and cannot save energy for O-Cloud resources can be solved through the above
  • the method described above configures the energy-saving policy (i.e., the first information) according to the O-Cloud resource occupancy (i.e., the measurement information) and sends it to O-Cloud, thereby realizing dynamic adjustment of O-Cloud resources and helping to reduce the overall energy consumption of O-Cloud. .
  • FIG. 5 is a schematic flowchart of a method for generating an energy-saving policy provided by an application embodiment of the present disclosure; as shown in Figure 5, the energy-saving policy can be generated by the Non-RT RIC/rApp in SMO.
  • the method includes:
  • Step 501 O-Cloud IMS provides Federated O-Cloud Orchestration & Management (FOCOM) in SMO through the O2ims interface, and O-Cloud DMS provides Network Function Orchestration (Network Function Orchestration) in SMO through the O2dms interface. NFO) reports the second measurement information;
  • FOCOM Federated O-Cloud Orchestration & Management
  • NFO Network Function Orchestration
  • the second measurement information is the same as the first measurement information in the method shown in Figure 4.
  • Table 2 For details of the second measurement information, see Table 2, which will not be described again here.
  • Reporting methods include periodic reporting and/or event-triggered reporting.
  • Step 502 FOCOM reports the measurement information of O-Cloud IMS to Non-RT RIC/rApp, and NFO reports the measurement information of O-Cloud DMS to Non-RT RIC/rApp.
  • Step 503 Non-RT RIC/rApp generates an energy-saving policy based on the second measurement information reported by O-Cloud IMS and/or O-Cloud DMS, which at least needs to include O-Cloud's energy-saving policy;
  • Non-RT RIC/rApp can also be combined with measurement information reported by VNF and/or CNF through the O1 interface, such as network business load, such as the number of users, the number of PRBs, PRB utilization, etc.
  • NFO NFO
  • Step 505 Based on the received energy-saving policy, NFO delivers the second configuration information to O-Cloud DMS through the O2dms interface for O-Cloud to perform energy-saving configuration.
  • a deployment management request may be sent, and the deployment management request carries the second configuration information.
  • the second configuration information may include but is not limited to at least one of the following: CPU frequency mode or CPU frequency option, second information, number of virtual CPUs, CPU power status and other parameters.
  • the second configuration information is the same as the first configuration information in the method shown in Figure 4. For details, see Table 3; no further details will be given here.
  • Step 506 O-Cloud DMS updates VNF and/or CNF deployment according to the energy-saving policy configured by NFO;
  • Step 507 O-Cloud DMS sends a success or failure response to NFO after updating.
  • Non-RT RIC/rApp receives the measurement information reported by O-Cloud IMS and sends it through FOCOM. , generate O-Cloud's energy-saving policy, generate second configuration information based on the energy-saving policy through NFO, and send the second configuration information to O-Cloud DMS.
  • O-Cloud's VNF and/or CNF-related energy-saving policies can be pre-configured inside the SMO, as well as related trigger conditions, thresholds, etc. required for CPU configuration updates.
  • SMO After receiving the measurement information reported by O-Cloud IMS or O-Cloud DMS, SMO determines the CPU configuration information based on the energy-saving policy, and sends the configuration information to O-Cloud DMS through NFO.
  • the energy-saving method for cloud platforms enhances the O2dms interface to deliver configuration parameters such as CPU frequency to O-Cloud DMS to facilitate O-Cloud DMS to implement energy-saving strategies.
  • SMO calculation obtains the current O -Cloud power consumption, generates energy-saving policies and sends them to O-Cloud DMS through the O2dms interface, including CPU resource adjustment parameters, such as CPU frequency, CPU status, etc., and the corresponding O-Cloud DMS dynamically adjusts O-Cloud CPU and other resources, with Helps reduce O-Cloud's overall energy consumption.
  • FIG. 6 is a schematic flowchart of another method of generating an energy-saving policy provided by an application embodiment of the present disclosure; as shown in Figure 6, a method of generating an O-Cloud energy-saving policy based on a Cloud Operator is provided.
  • Cloud Operator Receive the measurement information reported by O-Cloud IMS sent through SMO's FOCOM, and/or receive the measurement information reported by O-Cloud DMS sent through NFO, generate the O-Cloud energy-saving policy, and generate a third party based on the energy-saving policy through NFO Configuration information, and send third configuration information to O-Cloud DMS.
  • the methods include:
  • Step 601 O-Cloud IMS reports the current third measurement information to FOCOM in SMO through the O2ims interface, and O-Cloud DMS reports the current third measurement information to the NFO in SMO through the O2dms interface;
  • the third measurement information includes but is not limited to at least one of the following:
  • the third information is used to indicate the binding relationship between VNF and/or CNF and the currently bound CPU core
  • the resource usage of the CPU core bound to the current VNF and/or CNF (including CPU frequency, CPU load, CPU utilization, etc.).
  • the network function is VNF and/or CNF as an example.
  • the network function It can also include Near-RT RIC, O-CU, O-CU-CP, O-CU-UP, O-DU, O-RU, etc.
  • the third measurement information is the same as the above-mentioned first measurement information and second measurement information. For details, see Table 2, which will not be described again here. Reporting methods include periodic reporting and/or event-triggered reporting.
  • Step 602 FOCOM reports the measurement information of O-Cloud IMS to Cloud Operator, and NFO reports the measurement information of O-Cloud DMS to Cloud Operator.
  • Step 603 Cloud Operator generates an energy-saving policy based on the third measurement information reported by O-Cloud IMS and/or O-Cloud DMS, which at least needs to include O-Cloud's energy-saving policy.
  • Cloud Operator can also combine the measurement information reported by VNF and/or CNF through the O1 interface, such as network business load, such as the number of users, the number of PRBs, PRB utilization, etc.
  • Step 604 Cloud Operator issues O-Cloud’s energy-saving strategy to NFO;
  • Step 605 Based on the received energy-saving policy, NFO delivers the third configuration information to O-Cloud DMS through the O2dms interface for O-Cloud to perform energy-saving configuration.
  • the third configuration information includes but is not limited to at least one of the following: CPU frequency mode or CPU frequency option, second information, number of virtual CPUs, CPU power status and other parameters.
  • the third configuration information is the same as the above-mentioned first configuration information and second configuration information. For details, see Table 3, which will not be described again here.
  • Step 606 O-Cloud DMS updates VNF and/or CNF deployment according to the configuration information issued by NFO;
  • Step 607 O-Cloud DMS sends a success or failure response to NFO after updating.
  • SMO generates an energy-saving policy, which includes related configuration parameters of CPU frequency. Specifically, after receiving the data reported based on O-Cloud IMS and/or O-Cloud DMS, SMO determines the binding status of VNF and/or CNF (i.e. network function) to the CPU core, for example, the first VNF/CNF
  • VNF and/or CNF i.e. network function
  • the bound CPU cores (assumed to be CPU core x, CPU core y, CPU core z), and the CPU utilization determines whether to downclock the VNF/CNF bound CPU core, for example, by setting one or a group of gates Limit triggers frequency reduction.
  • one way to implement it is to set the threshold value t1 to determine that the CPU utilization (cpuUtilization) of the CPU core x, CPU core y, and CPU core z bound to the first VNF/CNF is less than t1, and then set the CPU core x, CPU core y, and CPU core z.
  • CPU core y and CPU core z execute CPU frequency reduction and configure the frequency (cupFreqSelect) Update to energy-saving mode (powersave), as shown in Figure 7.
  • another implementation method is to determine the CPU core x bound to the first VNF/CNF by setting a set of threshold values, such as t1, t2,...,tn (t1 ⁇ t2 ⁇ ... ⁇ tn) , the utilization rates of CPU core y and CPU core z are respectively lower than the corresponding threshold values (assumed to be t1, t2, t3), perform CPU downclocking on CPU core x, CPU core y, and CPU core z, and update the frequency configuration.
  • thresholds such as f1, f2,...,fn (f1 ⁇ f2 ⁇ ... ⁇ fn); as shown in Figure 8.
  • SMO generates an energy-saving policy, where the energy-saving policy includes relevant configuration parameters of the CPU core. Specifically, after receiving the data reported based on O-Cloud IMS and/or O-Cloud DMS, SMO determines the binding status of VNF/CNF and CPU core, for example, the CPU core bound to the second VNF/CNF (assuming Determine whether to modify the CPU core bound to the second VNF/CNF for CPU core x, CPU core y, CPU core z), and CPU utilization, for example, trigger the release of a CPU core by setting the CPU utilization threshold value .
  • One implementation method is to determine the utilization of CPU core x, CPU core y, and CPU core z bound to the second VNF/CNF by setting threshold values t1 and t2. For example, when the utilization of CPU core z is lower than t1 , and when the utilization rate of CPU core x and CPU core y is not higher than t2, send configuration information to O-Cloud DMS through O2dms, modify the CPU core binding relationship of the second VNF/CNF, and update it to CPU core x and CPU core y, thereby releasing the CPU core z.
  • the CPU core z can be further shut down to achieve energy saving, as shown in Figure 9.
  • SMO generates an energy saving policy that includes relevant configuration parameters of the CPU power state. Specifically, after receiving the data reported based on O-Cloud IMS and/or O-Cloud DMS, SMO will determine the binding status of VNF/CNF and CPU core. For example, when the physical node only deploys the third VNF/CNF or the It is determined that the CPU load used by other VNF/CNF is much lower than that used by the third VNF/CNF.
  • the CPU cores bound to the third VNF/CNF include CPU core x, CPU core y, CPU core z, and the CPU cores of each CPU core.
  • CPU utilization determines whether to modify the power configuration of the physical node where the CPU core bound to the third VNF/CNF is located. For example, setting the CPU utilization threshold value triggers modification of the CPU power configuration of the physical node where the core is located.
  • one implementation method is to determine the utilization of CPU core x, CPU core y, and CPU core z bound to the third VNF/CNF by setting the threshold value t1. For example, when the utilization of CPU core x is lower than t1 At this time, the configuration information is sent to O-Cloud DMS through O2dms, and the CPU power configuration of the physical node where the third VNF/CNF is located is modified from State1 to State2. For example, change the physical node from C0/P0 in G0 working mode to C0/P1 or C1-C7, or directly change the G0 working mode to G1-G3, as shown in Figure 10.
  • the priority of resource adjustment policy parameters involved in the energy-saving policy can be set.
  • SMO gives priority to adjusting the CPU frequency or adjusting the number of CPU cores used when generating energy-saving policies, and finally modifies the CPU power state.
  • Figure 11 is a schematic structural diagram of a communication device provided by an embodiment of the present disclosure; as shown in Figure 11, the device is applied to a network element, and the device includes:
  • the first sending module is configured to send first information to the cloud platform; the first information is used to indicate the computing resource configuration of the network function of the cloud platform, and the first information is determined based on at least the measurement information of the cloud platform.
  • the first information includes at least one of the following:
  • the second information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the second information includes at least one of the following:
  • the device further includes:
  • a first receiving module configured to receive measurement information from the cloud platform
  • a first determination module configured to determine the first information based on the measurement information
  • the measurement information includes at least one of the following:
  • the third information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the third information includes at least one of the following:
  • the network function includes at least one of the following:
  • the first determination module is used to perform at least one of the following:
  • the first rule includes at least one of the following:
  • the measurement information includes a second trigger condition corresponding to each parameter value combination and a second operation corresponding to the second trigger condition.
  • the first trigger condition includes: the parameter value belongs to the first threshold interval;
  • the second trigger condition includes: each parameter value in the parameter value combination belongs to the second threshold interval corresponding to the parameter value.
  • the first operation and the second operation include at least one of the following:
  • the network element includes at least one of the following: SMO, NFVO, MEO, VNFM, and CNFM.
  • the communication device provided in the above embodiment implements the corresponding communication method
  • only the division of the above program modules is used as an example.
  • the above processing can be assigned to be completed by different program modules as needed. That is, the internal structure of the network element is divided into different program modules to complete all or part of the processing described above.
  • the device provided by the above embodiments and the corresponding method embodiments belong to the same concept. Please refer to the method embodiments for the specific implementation process, which will not be described again here.
  • Figure 12 is a schematic structural diagram of another communication device provided by an embodiment of the present disclosure; as shown in Figure 12, the device is applied to a cloud platform, and the device includes:
  • the second receiving module is configured to receive the first information from the network element; the first information is used to indicate the computing resource configuration of the network function of the cloud platform, and the first information is determined based on at least the measurement information of the cloud platform .
  • the first information includes at least one of the following:
  • the second information is at least used to indicate the binding relationship between the network function of the cloud platform and the CPU core;
  • the second information includes at least one of the following:
  • the device further includes: a second sending module, configured to send measurement information to the network element;
  • the measurement information includes at least one of the following:
  • the third information is at least used to indicate the binding relationship between the virtual resources of the cloud platform and the CPU core;
  • the third information includes at least one of the following:
  • the network function includes at least one of the following:
  • the device further includes: a configuration module configured to configure computing resources of the cloud platform network function according to the first information.
  • the configuration module is used to perform at least one of the following:
  • the network element includes at least one of the following: SMO, NFVO, MEO, VNFM, CNFM.
  • the communication device provided in the above embodiment implements the corresponding communication method
  • only the division of the above program modules is used as an example.
  • the above processing can be assigned to be completed by different program modules as needed. That is, the internal structure of the cloud platform is divided into different program modules to complete all or part of the processing described above.
  • the device provided by the above embodiments and the corresponding method embodiments belong to the same concept. Please refer to the method embodiments for the specific implementation process, which will not be described again here.
  • Figure 13 is a schematic structural diagram of a communication device provided by an embodiment of the present disclosure.
  • the communication device 130 includes: a processor 1301 and a memory for storing computer programs that can run on the processor. 1302;
  • the communication device When the communication device is a network element, when the processor 1301 is used to run the computer program, perform: sending first information to the cloud platform; the first information is used to indicate the computing resource configuration of the cloud platform network function, The first information is determined based on at least measurement information of the cloud platform.
  • the communication device can also perform the method shown in Figure 2, which belongs to the same concept as the communication method embodiment shown in Figure 2. The specific implementation process can be found in the method embodiment, and will not be described again here.
  • the communication device When the communication device is a cloud platform, when the processor 1301 is used to run the computer program, perform: receiving first information from a network element; the first information is used to indicate the calculation of network functions of the cloud platform Resource configuration, the first information is determined based on at least the measurement information of the cloud platform.
  • the communication device can also perform the method shown in Figure 3, which belongs to the same concept as the communication method embodiment shown in Figure 3. The specific implementation process can be found in the method embodiment, and will not be described again here.
  • the communication device 130 may also include: at least one network interface 1303.
  • the various components in the communications device 130 are coupled together through a bus system 1304 .
  • the bus system 1304 is used to implement connection communication between these components.
  • the bus system 1304 also includes a power bus, a control bus, and a status signal bus.
  • the various buses are labeled as bus system 1304 in FIG. 13 .
  • the number of processors 1301 may be at least one.
  • the network interface 1303 is used for wired or wireless communication between the communication device 130 and other devices.
  • the memory 1302 in the embodiment of the present disclosure is used to store various types of data to support the operation of the communication device 130 .
  • the methods disclosed in the above embodiments of the present disclosure can be applied to the processor 1301 or implemented by the processor 1301.
  • the processor 1301 may be an integrated circuit chip with signal processing capabilities. During the implementation process, each step of the above method can be completed by instructions in the form of hardware integrated logic circuits or software in the processor 1301 .
  • the above-mentioned processor 1301 may be a general-purpose processor, a digital signal processor (DiGital Signal Processor, DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • DSP digital signal processor
  • the processor 1301 can implement or execute the disclosed methods, steps and logical block diagrams in the embodiments of the present disclosure.
  • a general-purpose processor may be a microprocessor or any conventional processor, etc.
  • the steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly implemented by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor.
  • the software module may be located in a storage medium, and the storage medium is located in the memory 1302.
  • the processor 1301 reads the information in the memory 1302, and completes the steps of the foregoing method in combination with its hardware.
  • the communication device 130 may be configured by one or more application specific integrated circuits (Application Specific Integrated Circuits, ASICs), DSPs, programmable logic devices (Programmable Logic Devices, PLDs), complex programmable logic devices (Complex Programmable Logic Device (CPLD), Field-Programmable Gate Array (FPGA), general-purpose processor, controller, microcontroller (Micro Controller Unit, MCU), microprocessor (Microprocessor), or other electronic Component implementation, used to execute the aforementioned methods.
  • ASICs Application Specific Integrated Circuits
  • DSPs digital signal processor
  • PLDs programmable logic devices
  • CPLD Complex Programmable Logic Device
  • FPGA Field-Programmable Gate Array
  • controller microcontroller
  • MCU Micro Controller Unit
  • MCU microprocessor
  • Microprocessor Microprocessor
  • Embodiments of the present disclosure also provide a computer-readable storage medium on which a computer program is stored;
  • the computer program When the computer-readable storage medium is applied to a network element, when the computer program is run by a processor, it executes: sending first information to the cloud platform; the first information is used to indicate the computing resource configuration of the cloud platform network function, The first information is determined based on at least measurement information of the cloud platform.
  • the computer program can also execute the method shown in Figure 2, which is of the same concept as the communication method embodiment shown in Figure 2. The specific implementation process can be found in the method embodiment, and will not be described again here.
  • the computer program when the computer-readable storage medium is applied to a cloud platform, when the computer program is run by a processor, it executes: receiving first information from a network element; the first information is used to indicate the calculation of network functions of the cloud platform. Resource configuration, the first information is determined based on at least the measurement information of the cloud platform Certainly. Specifically, the computer program can also execute the method shown in Figure 3, which belongs to the same concept as the communication method embodiment shown in Figure 3. The specific implementation process can be found in the method embodiment, and will not be described again here.
  • the disclosed devices and methods can be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division.
  • the coupling, direct coupling, or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the devices or units may be electrical, mechanical, or other forms. of.
  • the units described above as separate components may or may not be physically separated.
  • the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units; Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present disclosure can be all integrated into one processing unit, or each unit can be separately used as a unit, or two or more units can be integrated into one unit; the above-mentioned integration
  • the unit can be implemented in the form of hardware or in the form of hardware plus software functional units.
  • the aforementioned program can be stored in a computer-readable storage medium.
  • the program When the program is executed, Includes the steps of the above method embodiment; and the aforementioned storage media includes: mobile storage devices, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disks or optical disks, etc.
  • the above-mentioned integrated units of the present disclosure are implemented in the form of software function modules and sold or used as independent products, they can also be stored in a computer-readable storage medium.
  • the computer software products are stored in a storage medium and include a number of instructions to enables a computer device (which can be a personal computer, server, or network equipment, etc.) to perform all or part of the methods described in various embodiments of the present disclosure.
  • the aforementioned storage media include: mobile storage devices, ROM, RAM, magnetic disks or optical disks and other media that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Power Sources (AREA)

Abstract

本公开提供了一种通信方法、装置、设备和存储介质；所述方法包括：网元向云平台发送第一信息；所述第一信息用于指示云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。

Description

一种通信方法、装置、设备和存储介质
相关申请的交叉引用
本申请主张在2022年08月26日在中国提交的中国专利申请No.202211035703.9的优先权，其全部内容通过引用包含于此。
技术领域
本公开涉及无线通信领域，尤其涉及一种通信方法、装置、设备和存储介质。
背景技术
相关技术中，X86、高级精简指令集处理器(Advanced RISC Machines，ARM)等操作系统具备中央处理器(Central processing unit，CPU)频率调整功能，可以通过系统配置CPU扩展最大频率或最小频率，从而实现根据CPU利用率、CPU负载等状态调整CPU频率。
开放式无线电接入网(Open Radio Access Network，O-RAN)架构中定义云虚拟化平台(O-Cloud)，用于将通用物理资源经云化或虚拟化技术后部署无线网络功能，包括O-Cloud基础架构管理服务(Infrastructure Management Services，IMS)、O-Cloud部署管理服务(Deployment Management Services，DMS)、虚拟化网络功能(Virtualised Network Function，VNF)和/或容器化网络功能(Cloud Native Network Functions，CNF)。同时，O-Cloud可以通过O2接口与服务管理和编排(Service Management and Orchestration，SMO)交互，从而SMO对O-Cloud基础设施、VNF和/或CNF等资源进行管理。其中，SMO通过O2ims接口进行基础设施与云平台部署和资源管理，SMO通过O2dms接口进行VNF和/或CNF的部署和资源管理。
相关技术中，基于O-RAN架构部署VNF和/或CNF时，无法根据O-Cloud资源占用情况灵活配置并执行节能策略。
发明内容
有鉴于此，本公开的主要目的在于提供一种通信方法、装置、设备和存储介质。
为达到上述目的，本公开的技术方案是这样实现的：
本公开实施例提供了一种通信方法，应用于网元，所述方法包括：
向云平台发送第一信息；所述第一信息用于指示云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
上述方案中，所述第一信息包括以下至少之一：
第二信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU的频率选项或CPU的频率模式；
CPU的电源状态。
上述方案中，所述第二信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
CPU核的绑定规则；
虚拟CPU与CPU核的对应关系。
上述方案中，所述方法还包括：
接收来自所述云平台的测量信息；
基于所述测量信息，确定所述第一信息；
其中，所述测量信息包括以下至少之一：
第三信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU频率；
CPU利用率；
CPU的温度；
CPU的电源状态。
上述方案中，所述第三信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
虚拟CPU与CPU核的对应关系。
上述方案中，所述网络功能包括以下至少之一：
VNF；
CNF；
近实时无线智能控制器(Near-RT RIC)；
开放无线接入网控制单元(O-CU)；
开放无线接入网控制单元控制平面(O-CU-CP)；
开放无线接入网控制单元用户平面(O-CU-UP)；
开放无线接入网分布式单元(O-DU)；
开放无线接入网无线电单元(O-RU)。
上述方案中，所述基于所述测量信息，确定所述第一信息，包括以下至少之一：
运用预设的第一模型识别所述测量信息，确定第一信息；
基于预设的第一规则和所述测量信息，确定第一信息；
其中，所述第一规则包括以下至少之一：
所述测量信息中的每个参数值对应的第一触发条件和所述第一触发条件对应的第一操作；
所述测量信息包括的每种参数值组合对应的第二触发条件和所述第二触发条件对应的第二操作。
上述方案中，所述第一触发条件包括：参数值属于第一阈值区间；
所述第二触发条件包括：参数值组合中的每个参数值属于所述参数值相应的第二阈值区间。
上述方案中，所述第一操作和所述第二操作，包括以下至少之一：
调整CPU的频率选项或CPU的频率模式；
调整CPU的电源状态；
调整网络功能绑定的CPU核；
调整CPU核的绑定规则；
调整虚拟CPU与CPU核的对应关系。
上述方案中，所述网元包括以下至少之一：服务管理和编排(Service  Management and Orchestration，SMO)、网络功能虚拟化编排器(Network Functions Virtualisation Orchestrator，NFVO)、移动边缘协调(ME orchestrator，MEO)、虚拟网络功能管理器(Virtualised Network Function Manager，VNFM)、容器网络功能管理器(Container Network Function Manager，CNFM)。
本公开实施例提供了一种通信方法，应用于云平台，所述方法包括：
接收来自网元的第一信息；所述第一信息用于指示所述云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
上述方案中，所述第一信息包括以下至少之一：
第二信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU的频率选项或CPU的频率模式；
CPU的电源状态。
上述方案中，所述第二信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
CPU核的绑定规则；
虚拟CPU与CPU核的对应关系。
上述方案中，所述方法还包括：
向所述网元发送测量信息；
其中，所述测量信息包括以下至少之一：
第三信息，至少用于指示所述云平台的虚拟资源与CPU核的绑定关系；
虚拟CPU的数量；
CPU频率；
CPU利用率；
CPU的温度；
CPU的电源状态。
上述方案中，所述第三信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
虚拟CPU与CPU核的对应关系。
上述方案中，所述网络功能包括以下至少之一：
VNF；
CNF；
Near-RT RIC；
O-CU；
O-CU CP；
O-CU UP；
O-DU；
O-RU。
上述方案中，所述方法还包括：
根据所述第一信息进行云平台网络功能的计算资源的配置。
上述方案中，根据所述第一信息进行云平台网络功能的计算资源的配置，包括以下至少之一：
调整CPU的频率选项或CPU的频率模式；
调整CPU的电源状态；
调整网络功能绑定的CPU核；
调整CPU核的绑定规则；
调整虚拟CPU与CPU核的对应关系。
上述方案中，所述网元包括以下至少之一：SMO、NFVO、MEO、VNFM、CNFM。
本公开实施例提供了一种通信装置，应用于网元，所述装置包括：第一发送模块，用于向云平台发送第一信息；所述第一信息用于指示云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
上述方案中，所述第一信息包括以下至少之一：
第二信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU的频率选项或CPU的频率模式；
CPU的电源状态。
上述方案中，所述第二信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
CPU核的绑定规则；
虚拟CPU与CPU核的对应关系。
上述方案中，所述装置还包括：
第一接收模块，用于接收来自所述云平台的测量信息；
第一确定模块，用于基于所述测量信息，确定所述第一信息；
所述测量信息包括以下至少之一：
第三信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU频率；
CPU利用率；
CPU的温度；
CPU的电源状态。
上述方案中，所述第三信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
虚拟CPU与CPU核的对应关系。
上述方案中，所述网络功能包括以下至少之一：
VNF；
CNF；
Near-RT RIC；
O-CU；
O-CU CP；
O-CU UP；
O-DU；
O-RU。
上述方案中，所述第一确定模块，用于执行以下至少之一：
运用预设的第一模型识别所述测量信息，确定所述第一信息；
基于预设的第一规则和所述测量信息，确定第一信息；
其中，所述第一规则包括以下至少之一：
所述测量信息中的每个参数值对应的第一触发条件和所述第一触发条件对应的第一操作；
所述测量信息包括的每种参数值组合对应的第二触发条件和所述第二触发条件对应的第二操作。
上述方案中，所述第一触发条件包括：参数值属于第一阈值区间；
所述第二触发条件包括：参数值组合中的每个参数值属于所述参数值相应的第二阈值区间。
上述方案中，所述第一操作和所述第二操作，包括以下至少之一：
调整CPU的频率选项或CPU的频率模式；
调整CPU的电源状态；
调整网络功能绑定的CPU核；
调整CPU核的绑定规则；
调整虚拟CPU与CPU核的对应关系。
上述方案中，所述网元包括以下至少之一：SMO、NFVO、MEO、VNFM、CNFM。
本公开实施例提供了一种通信装置，应用于云平台，所述装置包括：第二接收模块，用于接收来自网元的第一信息；所述第一信息用于指示所述云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
上述方案中，所述第一信息包括以下至少之一：
第二信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU的频率选项或CPU的频率模式；
CPU的电源状态。
上述方案中，所述第二信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
CPU核的绑定规则；
虚拟CPU与CPU核的对应关系。
上述方案中，所述装置还包括：第二发送模块，用于向所述网元发送测量信息；
所述测量信息包括以下至少之一：
第三信息，至少用于指示所述云平台的虚拟资源与CPU核的绑定关系；
虚拟CPU的数量；
CPU频率；
CPU利用率；
CPU的温度；
CPU的电源状态。
上述方案中，所述第三信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
虚拟CPU与CPU核的对应关系。
上述方案中，所述网络功能包括以下至少之一：
VNF；
CNF；
Near-RT RIC；
O-CU；
O-CU CP；
O-CU UP；
O-DU；
O-RU。
上述方案中，所述装置还包括：配置模块，用于根据所述第一信息进行云平台网络功能的计算资源的配置。
上述方案中，所述配置模块，用于执行以下至少之一：
调整CPU的频率选项或CPU的频率模式；
调整CPU的电源状态；
调整网络功能绑定的CPU核；
调整CPU核的绑定规则；
调整虚拟CPU与CPU核的对应关系。
上述方案中，所述网元包括以下至少之一：SMO、NFVO、MEO、VNFM、CNFM。
本公开实施例又提供了一种通信设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，其中，所述处理器执行所述程序时实现网元侧任一项所述方法的步骤；或者，所述处理器执行所述程序时实现云平台侧任一项所述方法的步骤。
本公开实施例还提供了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网元侧任一项所述方法的步骤；或者，所述计算机程序被处理器执行时实现云平台侧任一项所述方法的步骤。
本公开实施例所提供的一种通信方法、装置、设备和存储介质，所述方法包括：网元向云平台发送第一信息；所述第一信息用于指示云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定；云平台接收来自网元的第一信息。如此，网元可以根据云平台自身的测量信息确定第一信息，云平台基于第一信息进行网络功能的计算资源的灵活配置，实现针对云平台网络功能级别的计算资源调整以达到更精准的节能配置。
附图说明
图1为一种O-RAN架构的示意图；
图2为本公开实施例提供的一种通信方法的流程示意图；
图3为本公开实施例提供的另一种通信方法的流程示意图；
图4为本公开应用实施例提供的一种通信方法的流程示意图；
图5为本公开应用实施例提供的一种生成节能策略方法的流程示意图；
图6为本公开应用实施例提供的另一种生成节能策略方法的流程示意图；
图7为本公开应用实施例提供的一种CPU频率的配置策略生成方法的流 程示意图；
图8为本公开应用实施例提供的另一种CPU频率的配置策略生成方法的流程示意图；
图9为本公开应用实施例提供的一种CPU核的配置策略生成方法的流程示意图；
图10为本公开应用实施例提供的一种CPU电源状态的配置策略生成方法的流程示意图；
图11为本公开实施例提供的一种通信装置的结构示意图；
图12为本公开实施例提供的另一种通信装置的结构示意图；
图13为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
相关技术中，O-RAN架构如图1所示，O-Cloud可以通过O2接口与SMO交互。
在部署VNF和/或CNF时，SMO通过O2dms向O-Cloud DMS发送的配置文件中配置CPU内核的数量，从而触发O-Cloud DMS创建VNF和/或CNF所需的资源并实例化。但相关技术中SMO通过O2dms发给O-Cloud DMS的配置文件中仅指定虚拟CPU(vCPU)的数量，当部署的VNF和/或CNF当前工作负载较低时，O-Cloud需要相应减少为当前VNF和/或CNF分配的资源，例如：降低CPU频率、修改CPU电源状态等，而无法实现O-Cloud的CPU频率、CPU状态等资源的调整，也无法根据O-Cloud的资源占用情况进行灵活配置和/或节能操作。
基于此，本公开实施例提供一种通信方法，网元向云平台发送第一信息；所述第一信息用于指示云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定；云平台接收来自网元的第一信息。从而，实现云平台根据自身的测量信息进行网络功能的计算资源的灵活配置。
下面结合实施例对本公开再作进一步详细的说明。
图2为本公开实施例提供的一种通信方法的流程示意图；如图2所示，所述方法可以应用于网元，所述方法包括：
步骤201、向云平台发送第一信息；所述第一信息用于指示云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
这里，所述网元包括但不限于以下至少之一：服务管理和编排(Service Management and Orchestration，SMO)、网络功能虚拟化编排器(Network Functions Virtualisation Orchestrator，NFVO)、移动边缘协调(ME orchestrator，MEO)、虚拟网络功能管理器(Virtualised Network Function Manager，VNFM)、容器网络功能管理器(Container Network Function Manager，CNFM)。
所述网元也可以是一种网络设备、网络功能，其可以与云平台进行交互，例如，可以是SMO，或者是SMO中的某个功能单元。
在一些实施例中，所述计算资源配置至少包括：CPU配置。
在一些实施例中，所述第一信息包括但不限于以下至少之一：
第二信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU的频率选项或CPU的频率模式；
CPU的电源状态。
在一些实施例中，所述第二信息，包括但不限于以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
CPU核的绑定规则；
虚拟CPU与CPU核的对应关系。
具体的，所述网元通过第一信息指示云平台进行网络功能计算资源的配置，第一信息可以包括操作所需的相关参数，如以下至少之一：第二信息、虚拟CPU的数量、CPU的频率选项或CPU的频率模式、CPU的电源状态。云平台基于接收到的第一信息包括的参数进行相关的资源配置，从而达到节能等效果。
这里，云平台可以通过一个或多个CPU实现。
CPU为物理CPU，物理CPU具有一个或多个CPU核(也称CPU内核)，即CPU可以是单核CPU、多核CPU等，相应的CPU可以具有单个CPU内核、多个CPU内核。CPU上可以实现一个或多个虚拟CPU。
应用时，通过虚拟化技术，基于虚拟CPU与CPU核的映射关系，在网络功能绑定的CPU核上为网络功能分配相应的虚拟CPU。
虚拟CPU的数量可以是CPU为网络功能分配的虚拟CPU的数量。若云平台实现多个网络功能，则虚拟CPU的数量可以包括：针对每个网络功能分配的虚拟CPU的数量。
针对云平台(如O-Cloud)中的无线网络功能，如集中单元(Centralized Unit，CU)、分布单元(Distributed Unit，DU)、无线接入网智能控制器(RAN Intelligent Controller，RIC)等通过虚拟化技术实现相应软件功能，虚拟化无线网络功能可以部署在通用服务器上，这里称为网络功能。网络功能可以被分配物理服务器上特定的计算、存储、网络等资源，网络功能与已分配的资源共同构成实例。所述网络功能包括以下至少之一：
VNF；
CNF；
近实时无线智能控制器(Near-RT RIC)；
开放无线接入网控制单元(O-RAN Central Unit，O-CU)；
O-CU控制平面(O-CU Control Plane，O-CU-CP)；
O-CU用户平面(O-CU User Plane，O-CU-UP)；
开放无线接入网分布式单元(O-RAN Distributed Unit，O-DU)；
开放无线接入网无线电单元(O-RAN Radio Unit，O-RU)等。
其中，O-RAN是指开放无线接入网(Open Radio Access Network)。
CPU的频率选项或CPU的频率模式，可以包括但不限于以下之一：
performance(高性能模式，根据配置的最大频率执行)；
powersave(节能模式，根据配置的最小频率执行)；
userspace(根据用户空间设置的CPU频率执行，需要携带目标CPU频率配置参数，例如，缩放设置速度(scaling_set_speed))；
schedutil(内核空间调度器计算各任务负载，并由CPU频率控制器基于CPU利用率调整CPU频率)；
ondemand(根据CPU的负载按比例调整CPU频率)；
conservative(根据CPU的负载按照步长调整CPU频率)。
CPU的电源状态，可以包括但不限于以下之一：
G-State(全局系统级别的电源状态，其下对应其他状态)；
S-State(CPU休眠状态，数字越大休眠程度越高)；
C-State(仅对G0状态有效，CPU内核级别工作状态，数字越大节能程度越高)；
P-State(系统性能设定状态，通过调整CPU频率降低CPU能耗，P0为最高性能，P1、P2等为节省电能的状态)；
T-State(节能状态，通过调整CPU运行时间达到节能)。
CPU核的绑定规则，可以包括但不限于以下之一：STATIC、DYNAMIC。
其中，STATIC可以是指静态；DYNAMIC可以是指动态。
需要说明的是，CPU的频率选项或CPU的频率模式、CPU的电源状态、CPU核的绑定规则等可以基于实际应用场景预先设定或配置，以上描述不对其进行限定。
实际应用时，云平台可以将自身的状态告诉网元，由网元针对该云平台的状态提供节能策略和/或计算资源的调整策略，即即发送第一信息。
基于此，在一些实施例中，所述方法还包括：
接收来自所述云平台的测量信息；
基于所述测量信息，确定所述第一信息；
其中，所述测量信息包括但不限于以下至少之一：
第三信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU频率；
CPU利用率；
CPU的温度；
CPU的电源状态。
这里，云平台对自身状态进行测量，得到测量信息；将测量信息上报给网元，网元基于测量信息确定第一信息。
需要说明的是，所述CPU利用率可以指CPU核的利用率和/或物理CPU的利用率；所述CPU频率可以指CPU核的频率和/或物理CPU的频率；这里 不做限定。CPU的温度一般指物理CPU的温度；CPU的电源状态一般指物理CPU的电源状态。
在一些实施例中，所述第三信息，包括但不限于以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
虚拟CPU与CPU核的对应关系。
这里，云平台可以对自身的CPU频率、CPU利用率、CPU的温度、CPU的电源状态、当前运行的网络功能的情况(如网络功能是否支持CPU核绑定、网络功能绑定的CPU核、虚拟CPU与CPU核的对应关系)、虚拟CPU的数量(可以具体包括为每个网络功能分别分配的虚拟CPU的数量)等进行检测，得到测量信息，告知给网元后可以供网元生成节能策略。
实际应用时，云平台检测后得到测量信息，通过测量信息告知网元自身当前的状态；网元基于测量信息确定第一信息，将第一信息发送给云平台；云平台基于第一信息进行网络功能的计算资源的配置。例如，将某一网络功能与CPU核绑定、将CPU的频率选项或CPU的频率模式调整为高性能模式等。
另外，可以设定节能策略涉及到的资源调整策略参数的优先级。例如，在生成节能策略时优先调整CPU频率或调整使用的CPU核的数量，最后修改CPU的电源状态。
在一些实施例中，测量信息和/或配置信息，还可以包括集群相关标识信息。所述集群相关配置信息包括但不限于以下至少之一：
云标识(cloudId)，即云平台的标识符，用于标识当前网络功能所在的云平台；
第一标识，用于标识当前网络功能所在云平台集群的主节点；
第二标识，用于标识当前网络功能所在云平台集群的工作节点；
第三标识，用于标识当前网络功能的实例。
在一些实施例中，所述基于所述测量信息，确定所述第一信息，包括：
基于预设的第一规则和所述测量信息，确定所述第一信息；
运用预设的第一模型识别所述测量信息，确定所述第一信息。
这里，预设的第一规则和预设的第一模型可以是基于任意需求制定，对于第一规则和/或第一模型的制定方式、制定内容均不作限定。
在一些实施例中，所述第一规则包括但不限于以下至少之一：
所述测量信息中的每个参数值对应的第一触发条件和所述第一触发条件对应的第一操作；
所述测量信息包括的每种参数值组合对应的第二触发条件和所述第二触发条件对应的第二操作。
在一些实施例中，所述第一触发条件包括：参数值属于第一阈值区间；
所述第二触发条件包括：参数值组合中的每个参数值属于所述参数值相应的第二阈值区间。
在一些实施例中，所述第一操作和所述第二操作，包括但不限于以下至少之一：
调整CPU的频率选项或CPU的频率模式；
调整CPU的电源状态；
调整网络功能绑定的CPU核；
调整CPU核的绑定规则；
调整虚拟CPU与CPU核的对应关系。
具体的，所述测量信息包括的参数值，可以是测量信息中任一参数的实时的测量值或统计值；
参数值组合，可以包括测量信息中多个参数的实时的测量值或统计值。
举例来说，参数值可以包括以下至少之一参数的数值：CPU频率、CPU利用率、CPU的温度。
针对每个参数可以分别设置不同的第一触发条件，针对每种参数组合也可以设置不同的第二触发条件。具体可以针对每个参数对应设置不同的阈值区间。
例如，针对CPU频率，设置其对应的阈值区间，可以包括：一个或一组CPU利用率的阈值区间，如[0，f1)、[f1，f2)、……[fn-1，fn]；
针对CPU利用率，设置其对应的阈值区间，可以包括：一个或一组CPU利用率的阈值区间，如[0，t1)、[t1，t2)、……[tn-1，tn]；
针对CPU的温度，设置其对应的阈值区间，可以包括：一个或一组CPU的温度的阈值区间，如[0，p1)、[p1，p2)、……[pn-1，pn]。
应用时，根据云平台测量并发送的测量信息确定当前各个网络功能绑定的CPU核、虚拟CPU等信息，结合网络功能绑定的CPU的资源占用情况(如CPU频率所属的阈值区间、CPU利用率所属的阈值区间、CPU的温度所属的阈值区间)，确定是否需要调整和/或具体如何调整操作，即确定第一操作或第二操作。
提供一种示例，某个云平台的某一网络功能分配有CPU核1、CPU核2；根据测量信息确定CPU核1的CPU利用率属于阈值区间[t1，t2)、CPU核2的CPU利用率属于阈值区间[0，t1)(t2大于t1，t1为利用率最低阈值)，则可以确定操作为调整网络功能绑定的CPU核，例如是：解除CPU核2与网络功能的绑定关系，释放CPU核2。
提供再一种示例，某个云平台的某一网络功能分配有CPU核1；根据测量信息确定CPU核1的CPU利用率属于阈值区间[t2，t3)、CPU频率属于阈值区间[f1，f2)；则可以确定操作为调整CPU的频率选项或CPU的频率模式，例如是：从performance(高性能模式)调整为powersave(节能模式)，或者，调整为自定义模式(userspace模式)。
需要说明的是，以上阈值区间的设定、参数值组合、执行的操作，仅作为一种示例，对触发条件、阈值区间、相应的第一操作和第二操作不做限定。
上述操作通过第一信息告知云平台，实现对云平台计算资源的配置。
在一些实施例中，所述方法还包括：
根据训练样本集和神经网络模型进行训练，得到训练后的神经网络模型，作为所述第一模型；
其中，训练样本集包括多个训练样本，每个训练样本包括样本测量信息和标签；所述标签可以为样本第一信息。样本测量信息可以参照上述测量信息设置，样本第一信息可以参照上述第一信息设置，这里不多赘述。
在一些实施例中，所述接收来自所述云平台的测量信息，包括以下至少之一：
接收所述云平台按预设时间周期发送的测量信息；
接收所述云平台在满足预设条件时发送的测量信息。
其中，在所述云平台按预设时间周期发送测量信息时，所述测量信息包括以下至少之一：
预设时间周期内至少一个参数的测量值的平均值；
预设时间周期内至少一个参数的测量值的最大值；
预设时间周期内至少一个参数的测量值的最小值。
这里，所述至少一个参数可以包括：CPU频率、CPU利用率、CPU的温度、CPU的电源状态等。
其中，所述满足预设条件包括：
所述测量信息中至少一个参数的测量值和/或统计值超过预设阈值。
所述统计值可以是时间周期内的最大值、最小值、平均值、总值等，具体可以基于实际需求设定。
在一些实施例中，所述第一信息包括以下至少之一：
性能模式；
均衡模式；
功率管理模式或节能模式。
在一些实施例中，当所述第一信息包括性能模式时，所述第一信息用于指示所述云平台网络功能计算资源增加或保持不变，支持高性能运行；或，
当所述第一信息包括均衡模式时，所述第一信息用于指示所述云平台网络功能计算资源在预设范围内增加或减少，支持自适应调整；或，
当所述第一信息包括功率管理模式或节能模式时，所述第一信息用于指示所述云平台网络功能计算资源减少，支持节能。
在一些实施例中，用于所述云平台网络功能计算资源调整的计算资源调整参数包括如下至少之一：
虚拟CPU的数量；
CPU的频率选项或CPU的频率模式或CPU的性能状态；
CPU的功率状态或CPU的电源状态或CPU的空闲状态；
CPU的温度状态。
其中，所述云平台网络功能计算资源调整，可以包括如下至少一项：所 述云平台网络功能计算资源增加或保持不变，实现高性能运行；所述云平台网络功能计算资源在预设范围内增加或减少，实现自适应资源调整；所述云平台网络功能计算资源减少，实现节能。
相应地，本公开实施例还提供了一种云平台侧的通信方法。
图3为本公开实施例提供的另一种通信方法的流程示意图；如图3所示，所述方法可以应用于云平台，所述方法包括：
步骤301、接收来自网元的第一信息；所述第一信息用于指示所述云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
其中，所述网元包括以下至少之一：SMO、NFVO、MEO、VNFM、CNFM。
所述云平台可以为O-Cloud、VIM、CIM等云平台。
在一些实施例中，所述第一信息包括但不限于以下至少之一：
第二信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU的频率选项或CPU的频率模式；
CPU的电源状态。
在一些实施例中，所述第二信息，包括但不限于以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
CPU核的绑定规则；
虚拟CPU与CPU核的对应关系。
在一些实施例中，所述方法还包括：
向所述网元发送测量信息；
其中，所述测量信息包括但不限于以下至少之一：
第三信息，至少用于指示所述云平台的虚拟资源与CPU核的绑定关系；
虚拟CPU的数量；
CPU频率；
CPU利用率；
CPU的温度；
CPU的电源状态。
在一些实施例中，所述第三信息，包括但不限于以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
虚拟CPU与CPU核的对应关系。
在一些实施例中，所述网络功能包括但不限于以下至少之一：
VNF；
CNF；
Near-RT RIC；
O-CU；
O-CU-CP；
O-CU-UP；
O-DU；
O-RU。
以上第一信息、测量信息、网络功能等已在图2所示方法中说明，这里不多赘述。
实际应用时，云平台接收到第一信息后，根据第一信息的指示配置网络功能的计算资源。
基于此，在一些实施例中，所述方法还包括：
根据所述第一信息进行云平台网络功能的计算资源的配置；可以包括但不限于以下至少之一：
调整CPU的频率选项或CPU的频率模式；
调整CPU的电源状态；
调整网络功能绑定的CPU核；
调整CPU核的绑定规则；
调整虚拟CPU与CPU核的对应关系。
以上调整具体可以基于第一信息中包括的内容执行。
这里，云平台根据第一信息可以确定并指示以上至少一种操作。
在一些实施例中，所述向所述网元发送测量信息，包括以下至少之一：
按预设时间周期向所述网元发送测量信息；
在满足预设条件时，向所述网元发送测量信息。
在一些实施例中，所述测量信息包括以下至少之一：
预设时间周期内至少一个参数的测量值的平均值；
预设时间周期内至少一个参数的测量值的最大值；
预设时间周期内至少一个参数的测量值的最小值。
在一些实施例中，所述满足预设条件包括：
所述测量信息中至少一个参数的测量值和/或统计值超过预设阈值。
通过图2和图3所示的方法，网元根据云平台的测量信息，实现网络功能级别的、针对某一个或某几个CPU内核的调整(如CPU的频率选项或CPU的频率模式、CPU的电源状态、虚拟CPU与CPU核的对应关系等调整)，相较于针对物理服务器通过BIOS调整所有CPU内核的CPU频率的方法，具有更加细粒度、定制化的效果，减少多余的CPU能耗开销，从而达到云平台节能的目的。
本公开实施例还提供一种O-RAN架构，该架构包括：云平台、网元；所述云平台可以是O-Cloud，所述网元可以是SMO。SMO针对O-Cloud资源利用率较低的情况，可以结合网络功能负载情况生成并向O-Cloud DMS下发节能策略，从而使O-Cloud DMS动态调整网络功能绑定的CPU的配置，降低O-Cloud能耗，达到节能的目的。
图4为本公开应用实施例提供的一种节能方法的流程示意图；如图4所示，所述方法包括：
步骤401、O-Cloud IMS通过O2ims接口(401.1)、O-Cloud DMS通过O2dms接口(401.2)向SMO上报当前的第一测量信息。
其中，所述第一测量信息可以包括以下至少之一：
第三信息，用于指示VNF和/或CNF与当前绑定的CPU核的绑定关系；
虚拟CPU的数量；
VNF和/或CNF所绑定CPU核的资源占用情况(包括CPU频率、CPU的负载、CPU利用率、CPU的温度、CPU的电源状态等)。
其中，第三信息可以包括以下至少之一：VNF和/或CNF是否支持CPU 核绑定；VNF和/或CNF绑定的至少一个CPU核；虚拟CPU与CPU核的对应关系。
这里以网络功能为VNF和/或CNF为例进行说明，实际应用时，网络功能还可以包括Near-RT RIC、O-CU、O-CU-CP、O-CU-UP、O-DU、O-RU等。
上报方式包括：周期性上报和/或事件触发上报。
步骤402、SMO基于O-Cloud IMS和O-Cloud DMS上报的第一测量信息，生成节能策略(至少需要包括O-Cloud的节能策略)；
这里，SMO还可以结合VNF和/或CNF通过O1接口上报的测量信息(可以包括网络业务负载，如用户数、PRB数、PRB利用率等)，生成节能策略。
步骤403、基于节能策略，SMO通过O2dms接口向O-Cloud DMS下发第一配置信息；
所述第一配置信息用于指示O-Cloud执行节能配置，包括但不限于以下至少之一：CPU频率模式或CPU频率选项、第二信息、虚拟CPU的数量、CPU电源状态等参数；
第二信息包括以下至少之一：VNF和/或CNF是否支持CPU核绑定；VNF和/或CNF绑定的至少一个CPU核；CPU核的绑定规则；虚拟CPU与CPU核的对应关系。
步骤404、O-Cloud DMS根据SMO发送的第一配置信息，更新VNF和/或CNF的部署；
具体可以发送部署管理请求，所述部署管理请求携带第一配置信息。
步骤405、O-Cloud DMS更新后向SMO发送成功或失败响应。
根据上述流程步骤401、步骤403，通过O2ims接口和/或O2dms接口上报的第一测量信息、通过O2dms接口下发的第一配置信息，可以包括集群相关标识信息，如表1所示。
表1

其中，masterNodeId是指主节点标识；workerNodeId是指工作节点标识；instanceId是指实例标识。
根据上述流程步骤401，O-Cloud IMS通过O2ims接口以及O-Cloud DMS通过O2dms接口向SMO上报第一测量信息，以告知当前VNF和/或CNF的资源占用情况，上报的第一测量信息包括但不限于表2所示参数：
表2
根据步骤403，SMO通过O2dms接口向O-Cloud DMS下发的第一配置信息中的配置参数包括但不限于表3所示：
表3

其中，Bios是指基本输入输出系统(Basic Input Output System)。
其中，CPU频率模式或CPU频率选项，在操作系统运行中调整，可以包括但不限于以下之一：performance(高性能模式，根据配置的最大频率执行)；powersave(节能模式，根据配置的最小频率执行)；userspace(根据用户空间设置的CPU频率执行，需要携带目标CPU频率配置参数，例如，scaling_set_speed)；schedutil(内核空间调度器计算各任务负载，并由CPU频率控制器基于CPU利用率调整CPU频率)；ondemand(根据CPU的负载按比例调整CPU频率)；conservative(根据CPU的负载按照步长调整CPU频率)。
CPU的电源状态，可以包括但不限于以下之一：G-State(全局系统级别的电源状态，其下对应其他状态)；S-State(CPU休眠状态，数字越大休眠程度越高)；C-State(仅对G0状态有效，CPU内核级别工作状态，数字越大节能程度越高)；P-State(系统性能设定状态，通过调整CPU频率降低CPU能耗，P0为最高性能，P1、P2等为节省电能的状态)；T-State(节能状态，通过调整CPU运行时间达到节能)。
CPU核绑定策略，可以包括：STATIC，DYNAMIC等。
本公开实施例提出的云平台节能方法，针对O-Cloud节能方案，增强O2dms接口向O-Cloud DMS下发CPU频率等配置参数，以便于O-Cloud DMS执行节能策略，从而解决SMO通过O2dms接口向O-Cloud DMS下发的配置文件中仅指定vCPU数量而无法面向O-Cloud资源进行节能的问题，通过上 述方法实现了根据O-Cloud资源占用情况(即测量信息)配置节能策略(即第一信息)并发送给O-Cloud，实现O-Cloud资源动态调整，有助于降低O-Cloud整体能耗。
图5为本公开应用实施例提供的一种生成节能策略方法的流程示意图；如图5所示，可以由SMO中Non-RT RIC/rApp生成节能策略，所述方法包括：
步骤501、O-Cloud IMS通过O2ims接口向SMO中的O-Cloud编排和管理(Federated O-Cloud Orchestration&Management，FOCOM)，以及O-Cloud DMS通过O2dms接口向SMO中的网络功能编排(Network Function Orchestration，NFO)上报第二测量信息；
其中，第二测量信息与图4所示方法中第一测量信息相同，第二测量信息具体可以参见表2，这里不多赘述。
上报方式包括周期性上报和/或事件触发上报。
步骤502、FOCOM将O-Cloud IMS的测量信息上报给Non-RT RIC/rApp，NFO将O-Cloud DMS的测量信息上报给Non-RT RIC/rApp。
步骤503、Non-RT RIC/rApp基于O-Cloud IMS和/或O-Cloud DMS上报的第二测量信息，生成节能策略，至少需要包括O-Cloud的节能策略；
这里，Non-RT RIC/rApp还可以结合结合VNF和/或CNF通过O1接口上报的测量信息，例如网络业务负载，如用户数、PRB数、PRB利用率等。
步骤504、Non-RT RIC rApp向NFO下发O-Cloud节能策略；
步骤505、基于接收的节能策略，NFO通过O2dms接口向O-Cloud DMS下发第二配置信息，用于O-Cloud执行节能配置。
具体可以发送部署管理请求，所述部署管理请求携带第二配置信息。第二配置信息，可以包括但不限于以下至少之一：CPU频率模式或CPU频率选项、第二信息、虚拟CPU的数量、CPU电源状态等参数。
第二配置信息与图4所示方法中的第一配置信息相同，具体可以参见表3；这里不多赘述。
步骤506、O-Cloud DMS根据NFO配置的节能策略，更新VNF和/或CNF部署；
步骤507、O-Cloud DMS更新后向NFO发送成功或失败响应。
如图5所示，提供了一种由Non-RT RIC/rApp实现的O-Cloud节能策略的生成方法，由SMO中的Non-RT RIC/rApp接收通过FOCOM发送O-Cloud IMS上报的测量信息，生成O-Cloud的节能策略，并通过NFO基于节能策略生成第二配置信息，并向O-Cloud DMS发送第二配置信息。
在一些实施例中，所述SMO内部可以预先配置O-Cloud的VNF和/或CNF相关的节能策略，以及，涉及CPU配置更新所需相关的触发条件、门限等。SMO接收O-Cloud IMS或O-Cloud DMS上报的测量信息后，基于节能策略确定CPU配置信息，并通过NFO向O-Cloud DMS发送所述配置信息。
本公开实施例提供的针对云平台的节能方法，增强O2dms接口向O-Cloud DMS下发CPU频率等配置参数，便于O-Cloud DMS执行节能策略。具体的，基于O-Cloud IMS和/或O-Cloud DMS上报的VNF和/或CNF与CPU核的绑定情况以及CPU资源占用情况(如CPU利用率、CPU负载等)，SMO计算获得当前O-Cloud功耗，生成节能策略并通过O2dms接口下发给O-Cloud DMS，包括CPU资源调整参数，例如CPU频率、CPU状态等，相应的O-Cloud DMS动态调整O-Cloud CPU等资源，有助于降低O-Cloud整体能耗。
图6为本公开应用实施例提供的另一种生成节能策略方法的流程示意图；如图6所示，提供了一种基于云操作员(Cloud Operator)的O-Cloud节能策略生成方法，Cloud Operator接收通过SMO的FOCOM发送的O-Cloud IMS上报的测量信息，和/或，接收通过NFO发送的O-Cloud DMS上报的测量信息，生成O-Cloud节能策略，并通过NFO基于节能策略生成第三配置信息，并向O-Cloud DMS发送第三配置信息。所述方法包括：
步骤601、O-Cloud IMS通过O2ims接口向SMO中的FOCOM，以及O-Cloud DMS通过O2dms接口向SMO中的NFO上报当前第三测量信息；
所述第三测量信息包括但不限于以下至少之一：
第三信息，用于指示VNF和/或CNF与当前绑定的CPU核的绑定关系；
虚拟CPU的数量；
当前VNF和/或CNF所绑定CPU核的资源占用情况(包括CPU频率、CPU负载、CPU利用率等)。
这里以网络功能为VNF和/或CNF为例进行说明，实际应用时，网络功 能还可以包括Near-RT RIC、O-CU、O-CU-CP、O-CU-UP、O-DU、O-RU等。
所述第三测量信息与上述第一测量信息、第二测量信息相同，具体可以具体参见表2，这里不多赘述。上报方式包括周期性上报和/或事件触发上报。
步骤602、FOCOM将O-Cloud IMS的测量信息上报给Cloud Operator，NFO将O-Cloud DMS的测量信息上报给Cloud Operator。
步骤603、Cloud Operator基于O-Cloud IMS和/或O-Cloud DMS上报的第三测量信息，生成节能策略，至少需要包括O-Cloud的节能策略。
这里，Cloud Operator还可以结合VNF和/或CNF通过O1接口上报的测量信息，例如网络业务负载，如用户数、PRB数、PRB利用率等。
步骤604、Cloud Operator向NFO下发O-Cloud的节能策略；
步骤605、基于接收的节能策略，NFO通过O2dms接口向O-Cloud DMS下发第三配置信息，用于O-Cloud执行节能配置。
所述第三配置信息包括但不限于以下至少之一：CPU频率模式或CPU频率选项、第二信息、虚拟CPU的数量、CPU电源状态等参数。
所述第三配置信息与上述第一配置信息、第二配置信息相同，具体可以参见表3所示，这里不多赘述。
步骤606、O-Cloud DMS根据NFO下发的配置信息，更新VNF和/或CNF部署；
步骤607、O-Cloud DMS更新后向NFO发送成功或失败响应。
关于生成节能策略的方法，具体提供以下示例。
在一种应用实施例中，SMO生成节能策略，所述节能策略包括CPU频率的相关配置参数。具体来说，SMO在接收基于O-Cloud IMS和/或O-Cloud DMS上报的数据后，根据VNF和/或CNF(即网络功能)与CPU核的绑定情况，例如，第一VNF/CNF绑定的CPU核(假设为CPU核x、CPU核y、CPU核z)，以及，CPU利用率决定是否对VNF/CNF绑定的CPU核进行降频，例如，通过设置一个或一组门限值触发降频。
其中，一种实现方式是通过设置门限值t1判断第一VNF/CNF所绑定的CPU核x、CPU核y、CPU核z的CPU利用率(cpuUtilization)小于t1，并对CPU核x、CPU核y、CPU核z执行CPU降频，将频率配置(cupFreqSelect) 更新为节能模式(powersave)，如图7所示。
其中，另一种实现方式是通过设置一组门限值，例如t1,t2,...,tn(t1<t2<...<tn)判断第一VNF/CNF所绑定的CPU核x、CPU核y、CPU核z的利用率分别低于对应的门限值(假设为t1,t2,t3)，对CPU核x、CPU核y、CPU核z执行CPU降频，将频率配置更新为用户空间(userspace)模式，根据不同门限设置不同的目标CPU频率配置参数，例如f1,f2,...,fn(f1<f2<...<fn)；如图8所示。
在另一种实施例中，SMO生成节能策略，所述节能策略包括CPU核的相关配置参数。具体来说，SMO在接收基于O-Cloud IMS和/或O-Cloud DMS上报的数据后，根据VNF/CNF与CPU核的绑定情况，例如，第二VNF/CNF绑定的CPU核(假设为CPU核x、CPU核y、CPU核z)，以及CPU利用率决定是否对第二VNF/CNF绑定的CPU核进行修改，例如，通过设置CPU利用率门限值触发释放某个CPU核。
其中，一种实现方式是通过设置门限值t1与t2判断第二VNF/CNF所绑定的CPU核x、CPU核y、CPU核z的利用率，例如当CPU核z利用率低于t1，且CPU核x和CPU核y的利用率不高于t2时，通过O2dms向O-Cloud DMS发送配置信息，修改第二VNF/CNF的CPU核绑定关系，更新为CPU核x和CPU核y，从而释放CPU核z，可选的，可以进一步关闭CPU核z，从而达到节能的目的，如图9所示。
在再一种实施例中，SMO生成节能策略，所述节能策略包括CPU电源状态的相关配置参数。具体来说，SMO在接收基于O-Cloud IMS和/或O-Cloud DMS上报的数据后，根据VNF/CNF与CPU核的绑定情况，例如，当物理节点仅部署第三VNF/CNF或经判断其他VNF/CNF使用的CPU负载远低于第三VNF/CNF使用的CPU负载，第三VNF/CNF绑定的CPU核包括CPU核x、CPU核y、CPU核z，以及各CPU核的CPU利用率，决定是否对第三VNF/CNF绑定的CPU核所在物理节点电源配置进行修改，例如，通过设置CPU利用率门限值触发修改核所在物理节点CPU电源配置。
其中，一种实现方式是通过设置门限值t1判断第三VNF/CNF所绑定的CPU核x、CPU核y、CPU核z的利用率，例如当CPU核x利用率低于t1 时，通过O2dms向O-Cloud DMS发送配置信息，修改第三VNF/CNF所在物理节点CPU电源配置，从State1修改为State2。例如，将物理节点从G0工作模式下C0/P0修改为C0/P1或C1-C7，或者直接将G0工作模式修改为G1-G3，如图10所示。
实际应用时，可以设定节能策略涉及到的资源调整策略参数的优先级。例如，SMO在生成节能策略时优先调整CPU频率或调整使用的CPU核数量，最后修改CPU电源状态。
图11为本公开实施例提供的一种通信装置的结构示意图；如图11所示，所述装置应用于网元，所述装置包括：
第一发送模块，用于向云平台发送第一信息；所述第一信息用于指示云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
上述方案中，所述第一信息包括以下至少之一：
第二信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU的频率选项或CPU的频率模式；
CPU的电源状态。
上述方案中，所述第二信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
CPU核的绑定规则；
虚拟CPU与CPU核的对应关系。
上述方案中，所述装置还包括：
第一接收模块，用于接收来自所述云平台的测量信息；
第一确定模块，用于基于所述测量信息，确定所述第一信息；
其中，所述测量信息包括以下至少之一：
第三信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU频率；
CPU利用率；
CPU的温度；
CPU的电源状态。
上述方案中，所述第三信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
虚拟CPU与CPU核的对应关系。
上述方案中，所述网络功能包括以下至少之一：
VNF；
CNF；
Near-RT RIC；
O-CU；
O-CU CP；
O-CU UP；
O-DU；
O-RU。
上述方案中，所述第一确定模块，用于执行以下至少之一：
运用预设的第一模型识别所述测量信息，确定所述第一信息；
基于预设的第一规则和所述测量信息，确定第一信息；
其中，所述第一规则包括以下至少之一：
所述测量信息中的每个参数值对应的第一触发条件和所述第一触发条件对应的第一操作；
所述测量信息包括的每种参数值组合对应的第二触发条件和所述第二触发条件对应的第二操作。
上述方案中，所述第一触发条件包括：参数值属于第一阈值区间；
所述第二触发条件包括：参数值组合中的每个参数值属于所述参数值相应的第二阈值区间。
上述方案中，所述第一操作和所述第二操作，包括以下至少之一：
调整CPU的频率选项或CPU的频率模式；
调整CPU的电源状态；
调整网络功能绑定的CPU核；
调整CPU核的绑定规则；
调整虚拟CPU与CPU核的对应关系。
上述方案中，所述网元包括以下至少之一：SMO、NFVO、MEO、VNFM、CNFM。
需要说明的是：上述实施例提供的通信装置在实现相应通信方法时，仅以上述各程序模块的划分进行举例说明，实际应用中，可以根据需要而将上述处理分配由不同的程序模块完成，即将网元的内部结构划分成不同的程序模块，以完成以上描述的全部或者部分处理。另外，上述实施例提供的装置与相应方法的实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
图12为本公开实施例提供的另一种通信装置的结构示意图；如图12所示，所述装置应用于云平台，所述装置包括：
第二接收模块，用于接收来自网元的第一信息；所述第一信息用于指示所述云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
上述方案中，所述第一信息包括以下至少之一：
第二信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
虚拟CPU的数量；
CPU的频率选项或CPU的频率模式；
CPU的电源状态。
上述方案中，所述第二信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
CPU核的绑定规则；
虚拟CPU与CPU核的对应关系。
上述方案中，所述装置还包括：第二发送模块，用于向所述网元发送测量信息；
其中，所述测量信息包括以下至少之一：
第三信息，至少用于指示所述云平台的虚拟资源与CPU核的绑定关系；
虚拟CPU的数量；
CPU频率；
CPU利用率；
CPU的温度；
CPU的电源状态。
上述方案中，所述第三信息，包括以下至少之一：
网络功能是否支持CPU核绑定；
网络功能绑定的至少一个CPU核；
虚拟CPU与CPU核的对应关系。
上述方案中，所述网络功能包括以下至少之一：
VNF；
CNF；
Near-RT RIC；
O-CU；
O-CU CP；
O-CU UP；
O-DU；
O-RU。
上述方案中，所述装置还包括：配置模块，用于根据所述第一信息进行云平台网络功能的计算资源的配置。
上述方案中，所述配置模块，用于执行以下至少之一：
调整CPU的频率选项或CPU的频率模式；
调整CPU的电源状态；
调整网络功能绑定的CPU核；
调整CPU核的绑定规则；
调整虚拟CPU与CPU核的对应关系。
上述方案中，所述网元包括以下至少之一：SMO、NFVO、MEO、VNFM、 CNFM。
需要说明的是：上述实施例提供的通信装置在实现相应通信方法时，仅以上述各程序模块的划分进行举例说明，实际应用中，可以根据需要而将上述处理分配由不同的程序模块完成，即将云平台的内部结构划分成不同的程序模块，以完成以上描述的全部或者部分处理。另外，上述实施例提供的装置与相应方法的实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
图13为本公开实施例提供的一种通信设备的结构示意图，如图13所示，所述通信设备130包括：处理器1301和用于存储能够在所述处理器上运行的计算机程序的存储器1302；
所述通信设备为网元时，所述处理器1301用于运行所述计算机程序时，执行：向云平台发送第一信息；所述第一信息用于指示云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。具体来说，所述通信设备还可以执行如图2所示的方法，与图2所示的通信方法实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
所述通信设备为云平台时，所述处理器1301用于运行所述计算机程序时，执行：接收来自网元的第一信息；所述第一信息用于指示所述云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。具体来说，所述通信设备还可以执行如图3所示的方法，与图3所示的通信方法实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
实际应用时，所述通信设备130还可以包括：至少一个网络接口1303。所述通信设备130中的各个组件通过总线系统1304耦合在一起。可理解，总线系统1304用于实现这些组件之间的连接通信。总线系统1304除包括数据总线之外，还包括电源总线、控制总线和状态信号总线。但是为了清楚说明起见，在图13中将各种总线都标为总线系统1304。其中，所述处理器1301的个数可以为至少一个。网络接口1303用于通信设备130与其他设备之间有线或无线方式的通信。
本公开实施例中的存储器1302用于存储各种类型的数据以支持通信设备130的操作。
上述本公开实施例揭示的方法可以应用于处理器1301中，或者由处理器1301实现。处理器1301可能是一种集成电路芯片，具有信号的处理能力。在实现过程中，上述方法的各步骤可以通过处理器1301中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器1301可以是通用处理器、数字信号处理器(DiGital Signal Processor，DSP)，或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。处理器1301可以实现或者执行本公开实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者任何常规的处理器等。结合本公开实施例所公开的方法的步骤，可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于存储介质中，该存储介质位于存储器1302，处理器1301读取存储器1302中的信息，结合其硬件完成前述方法的步骤。
在示例性实施例中，通信设备130可以被一个或多个应用专用集成电路(Application Specific Integrated Circuit，ASIC)、DSP、可编程逻辑器件(Programmable Logic Device，PLD)、复杂可编程逻辑器件(Complex Programmable Logic Device，CPLD)、现场可编程门阵列(Field-Programmable Gate Array，FPGA)、通用处理器、控制器、微控制器(Micro Controller Unit，MCU)、微处理器(Microprocessor)、或其他电子元件实现，用于执行前述方法。
本公开实施例还提供了一种计算机可读存储介质，其上存储有计算机程序；
所述计算机可读存储介质应用于网元时，所述计算机程序被处理器运行时，执行：向云平台发送第一信息；所述第一信息用于指示云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。具体来说，所述计算机程序还可以执行如图2所示的方法，与图2所示的通信方法实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
所述计算机可读存储介质应用于云平台时，所述计算机程序被处理器运行时，执行：接收来自网元的第一信息；所述第一信息用于指示所述云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确 定。具体来说，所述计算机程序还可以执行如图3所示的方法，与图3所示的通信方法实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
在本公开所提供的几个实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。以上所描述的设备实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，如：多个单元或组件可以结合，或可以集成到另一个系统，或一些特征可以忽略，或不执行。另外，所显示或讨论的各组成部分相互之间的耦合、或直接耦合、或通信连接可以是通过一些接口，设备或单元的间接耦合或通信连接，可以是电性的、机械的或其它形式的。
上述作为分离部件说明的单元可以是、或也可以不是物理上分开的，作为单元显示的部件可以是、或也可以不是物理单元，即可以位于一个地方，也可以分布到多个网络单元上；可以根据实际的需要选择其中的部分或全部单元来实现本实施例方案的目的。
另外，在本公开各实施例中的各功能单元可以全部集成在一个处理单元中，也可以是各单元分别单独作为一个单元，也可以两个或两个以上单元集成在一个单元中；上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能单元的形式实现。
本领域普通技术人员可以理解：实现上述方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成，前述的程序可以存储于一个计算机可读取存储介质中，该程序在执行时，执行包括上述方法实施例的步骤；而前述的存储介质包括：移动存储设备、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
或者，本公开上述集成的单元如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。基于这样的理解，本公开实施例的技术方案本质上或者说对现有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中，包括若干指令用以使得一台计算机设备(可以是个人计算机、服务器、 或者网络设备等)执行本公开各个实施例所述方法的全部或部分。而前述的存储介质包括：移动存储设备、ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
需要说明的是：“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。
另外，本公开实施例所记载的技术方案之间，在不冲突的情况下，可以任意组合。
以上所述，仅为本公开的具体实施方式，但本公开的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本公开揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本公开的保护范围之内。因此，本公开的保护范围应以所述权利要求的保护范围为准。

Claims (26)

  1. 一种通信方法，应用于网元，所述方法包括：
    向云平台发送第一信息；所述第一信息用于指示云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
  2. 根据权利要求1所述的方法，其中，所述第一信息包括以下至少之一：
    第二信息，至少用于指示所述云平台的网络功能与中央处理器CPU核的绑定关系；
    虚拟CPU的数量；
    CPU的频率选项或CPU的频率模式；
    CPU的电源状态。
  3. 根据权利要求2所述的方法，其中，所述第二信息，包括以下至少之一：
    网络功能是否支持CPU核绑定；
    网络功能绑定的至少一个CPU核；
    CPU核的绑定规则；
    虚拟CPU与CPU核的对应关系。
  4. 根据权利要求1所述的方法，所述方法还包括：
    接收来自所述云平台的测量信息；
    基于所述测量信息，确定所述第一信息；
    其中，所述测量信息包括以下至少之一：
    第三信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
    虚拟CPU的数量；
    CPU频率；
    CPU利用率；
    CPU的温度；
    CPU的电源状态。
  5. 根据权利要求4所述的方法，其中，所述第三信息，包括以下至少之一：
    网络功能是否支持CPU核绑定；
    网络功能绑定的至少一个CPU核；
    虚拟CPU与CPU核的对应关系。
  6. 根据权利要求1至5任一项所述的方法，其中，所述网络功能包括以下至少之一：
    虚拟化网络功能VNF；
    容器化网络功能CNF；
    近实时无线智能控制器Near-RT RIC；
    开放无线接入网控制单元O-CU；
    开放无线接入网控制单元控制平面O-CU-CP；
    开放无线接入网控制单元用户平面O-CU-UP；
    开放无线接入网分布式单元O-DU；
    开放无线接入网无线电单元O-RU。
  7. 根据权利要求4所述的方法，其中，所述基于所述测量信息，确定所述第一信息，包括以下至少之一：
    运用预设的第一模型识别所述测量信息，确定所述第一信息；
    基于预设的第一规则和所述测量信息，确定所述第一信息；
    其中，所述第一规则包括以下至少之一：
    所述测量信息中的每个参数值对应的第一触发条件和所述第一触发条件对应的第一操作；
    所述测量信息包括的每种参数值组合对应的第二触发条件和所述第二触发条件对应的第二操作。
  8. 根据权利要求7所述的方法，其中，所述第一触发条件包括：参数值属于第一阈值区间；
    所述第二触发条件包括：参数值组合中的每个参数值属于所述参数值相应的第二阈值区间。
  9. 根据权利要求7所述的方法，其中，所述第一操作和所述第二操作，包括以下至少之一：
    调整CPU的频率选项或CPU的频率模式；
    调整CPU的电源状态；
    调整网络功能绑定的CPU核；
    调整CPU核的绑定规则；
    调整虚拟CPU与CPU核的对应关系。
  10. 根据权利要求1所述的方法，其中，所述网元包括以下至少之一：服务管理和编排SMO、网络功能虚拟化编排器NFVO、移动边缘协调MEO、虚拟网络功能管理器VNFM、容器网络功能管理器CNFM。
  11. 根据权利要求1所述的方法，其中，所述第一信息包括以下至少之一：
    性能模式；
    均衡模式；
    功率管理模式或节能模式。
  12. 根据权利要求11所述方法，其中，当所述第一信息包括性能模式时，所述第一信息用于指示所述云平台网络功能计算资源增加或保持不变，支持高性能运行；或，
    当所述第一信息包括均衡模式时，所述第一信息用于指示所述云平台网络功能计算资源在预设范围内增加或减少，支持自适应调整；或，
    当所述第一信息包括功率管理模式或节能模式时，所述第一信息用于指示所述云平台网络功能计算资源减少，支持节能。
  13. 根据权利要求12所述方法，其中，用于所述云平台网络功能计算资源调整的计算资源调整参数包括如下至少之一：
    虚拟CPU的数量；
    CPU的频率选项或CPU的频率模式或CPU的性能状态；
    CPU的功率状态或CPU的电源状态或CPU的空闲状态；
    CPU的温度状态。
  14. 一种通信方法，应用于云平台，所述方法包括：
    接收来自网元的第一信息；所述第一信息用于指示所述云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
  15. 根据权利要求14所述的方法，其中，所述第一信息包括以下至少之 一：
    第二信息，至少用于指示所述云平台的网络功能与CPU核的绑定关系；
    虚拟CPU的数量；
    CPU的频率选项或CPU的频率模式；
    CPU的电源状态。
  16. 根据权利要求15所述的方法，其中，所述第二信息，包括以下至少之一：
    网络功能是否支持CPU核绑定；
    网络功能绑定的至少一个CPU核；
    CPU核的绑定规则；
    虚拟CPU与CPU核的对应关系。
  17. 根据权利要求14所述的方法，所述方法还包括：
    向所述网元发送测量信息；
    其中，所述测量信息包括以下至少之一：
    第三信息，至少用于指示所述云平台的虚拟资源与CPU核的绑定关系；
    虚拟CPU的数量；
    CPU频率；
    CPU利用率；
    CPU的温度；
    CPU的电源状态。
  18. 根据权利要求17所述的方法，其中，所述第三信息，包括以下至少之一：
    网络功能是否支持CPU核绑定；
    网络功能绑定的至少一个CPU核；
    虚拟CPU与CPU核的对应关系。
  19. 根据权利要求14至18任一项所述的方法，其中，所述网络功能包括以下至少之一：
    VNF；
    CNF；
    Near-RT RIC；
    O-CU；
    O-CU CP；
    O-CU UP；
    O-DU；
    O-RU。
  20. 根据权利要求14所述的方法，所述方法还包括：
    根据所述第一信息进行云平台网络功能的计算资源的配置。
  21. 根据权利要求20所述的方法，其中，根据所述第一信息进行云平台网络功能的计算资源的配置，包括以下至少之一：
    调整CPU的频率选项或CPU的频率模式；
    调整CPU的电源状态；
    调整网络功能绑定的CPU核；
    调整CPU核的绑定规则；
    调整虚拟CPU与CPU核的对应关系。
  22. 根据权利要求14所述的方法，其中，所述网元包括以下至少之一：SMO、NFVO、MEO、VNFM、CNFM。
  23. 一种通信装置，应用于网元，所述装置包括：第一发送模块，用于向云平台发送第一信息；所述第一信息用于指示云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
  24. 一种通信装置，应用于云平台，所述装置包括：第二接收模块，用于接收来自网元的第一信息；所述第一信息用于指示所述云平台网络功能的计算资源配置，所述第一信息至少基于所述云平台的测量信息确定。
  25. 一种通信设备，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现权利要求1至13任一项所述方法的步骤；或者，所述处理器执行所述程序时实现权利要求14至22任一项所述方法的步骤。
  26. 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权利要求1至13任一项所述方法的步骤；或者，所述计算机程序被处理器执行时实现权利要求14至22任一项所述方法的步骤。
PCT/CN2023/113956 2022-08-26 2023-08-21 一种通信方法、装置、设备和存储介质 WO2024041472A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211035703.9 2022-08-26
CN202211035703.9A CN117675808A (zh) 2022-08-26 2022-08-26 一种通信方法、装置、设备和存储介质

Publications (1)

Publication Number Publication Date
WO2024041472A1 true WO2024041472A1 (zh) 2024-02-29

Family

ID=90012472

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/113956 WO2024041472A1 (zh) 2022-08-26 2023-08-21 一种通信方法、装置、设备和存储介质

Country Status (2)

Country Link
CN (1) CN117675808A (zh)
WO (1) WO2024041472A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200322226A1 (en) * 2019-04-02 2020-10-08 Sap Se Cloud resource scaling using programmable-network traffic statistics
CN112416513A (zh) * 2020-11-18 2021-02-26 烽火通信科技股份有限公司 一种云化网络下动态调节虚拟机主频的方法及系统
CN113301590A (zh) * 2021-05-24 2021-08-24 中科院计算所南京研究院 一种面向5g接入网的虚拟资源管控系统
CN113811013A (zh) * 2021-09-10 2021-12-17 阿里巴巴达摩院(杭州)科技有限公司 集群容量动态调节方法、设备、存储介质和通信系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200322226A1 (en) * 2019-04-02 2020-10-08 Sap Se Cloud resource scaling using programmable-network traffic statistics
CN112416513A (zh) * 2020-11-18 2021-02-26 烽火通信科技股份有限公司 一种云化网络下动态调节虚拟机主频的方法及系统
CN113301590A (zh) * 2021-05-24 2021-08-24 中科院计算所南京研究院 一种面向5g接入网的虚拟资源管控系统
CN113811013A (zh) * 2021-09-10 2021-12-17 阿里巴巴达摩院(杭州)科技有限公司 集群容量动态调节方法、设备、存储介质和通信系统

Also Published As

Publication number Publication date
CN117675808A (zh) 2024-03-08

Similar Documents

Publication Publication Date Title
NL2029059B1 (en) Hierarchical power management apparatus and method
US10613883B2 (en) Managing virtual machine migration
JP5981020B2 (ja) 効率的な消費電力管理のための動的インタラプト再コンフィグレーション
US9367340B2 (en) Resource management method and apparatus for virtual machine system, and virtual machine system
US11567556B2 (en) Platform slicing of central processing unit (CPU) resources
WO2017041556A1 (zh) 虚拟资源调度方法、装置及系统
US20200218676A1 (en) Adaptive processor resource utilization
US20230229495A1 (en) Task scheduling method and apparatus
EP3886385A1 (en) Device, system and method to dynamically prioritize a data flow based on user interest in a task
US9336106B2 (en) Dynamically limiting bios post for effective power management
CN102164044A (zh) 联网方法和联网系统
CN107203256B (zh) 一种网络功能虚拟化场景下的节能分配方法与装置
WO2024041472A1 (zh) 一种通信方法、装置、设备和存储介质
KR102333391B1 (ko) 전자 장치 및 이의 전력 제어 방법
Tesfatsion et al. PerfGreen: performance and energy aware resource provisioning for heterogeneous clouds
CN116868170A (zh) 嵌入式系统的运行方法和装置、嵌入式系统及芯片
US20090077290A1 (en) Controller for processing apparatus
US20210182194A1 (en) Processor unit resource exhaustion detection and remediation
US20220011843A1 (en) Software entity power consumption estimation and monitoring
Gvozdetska et al. Energy-efficient backfill-based scheduling approach for SLURM resource manager
US20230195201A1 (en) Coprocessor power management in hybrid architectures
US10887248B2 (en) IO bandwidth control method, IO access request processing method, apparatus, and system
WO2021109125A1 (zh) 一种弹性伸缩组的管理方法、装置
TWI488053B (zh) 雲端裝置配置方法
Otani et al. Accelerator-Resource Sharing with Offload-Mediated Proxy for vRAN

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: 23856562

Country of ref document: EP

Kind code of ref document: A1