Computing resource management method, system, and device

Abstract

Embodiments of the present application provide a computing resource management method, a system, and a device. The method comprises: receiving an instantiation request of a virtualized network function (VNF) from a network functions virtualization orchestrator (NFVO), wherein the instantiation request of the VNF comprises an operating mode of the VNF; determining addressing information of a heterogeneous computing resource pool and computing resource requirement information of the VNF which correspond to the operating mode of the VNF; and sending a computing resource allocation request to an infrastructure manager (IM), wherein the computing resource allocation request comprises the addressing information of the heterogeneous computing resource pool and the computing resource requirement information of the VNF, and the computing resource allocation request is used for requesting to allocate resources to the VNF from the heterogeneous computing resource pool on the basis of the addressing information of the heterogeneous computing resource pool and the computing resource requirement information of the VNF. The embodiments of the present application can meet the computing resource requirements of new telecommunication applications having high computing task concurrency.

Description

A method, system and device for managing computing resources

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office of China on February 8, 2024, with application number 202410177145.2, and invention name “A method, system and device for managing computing resources”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of communication technology, and in particular to a method, system, and device for managing computing resources. Background Art

[0003] Network functions virtualization (NFV) is a key technology for telecom network operators leveraging internet technologies to transform to the cloud. It decouples the software and hardware of telecom network functions, enabling the software-based implementation of these functions on general-purpose servers. With the rapid development of new telecom applications with high computational concurrency (such as 5G video, new call applications, AI, and converged computing and network applications), the processing power of the central processing units (CPUs) in network function virtualization infrastructure (NFVI) and the existing NFV architecture's approach to allocating computing resources to VNFs are struggling to meet the computing resource demands of these new telecom applications.

[0004] Therefore, how to meet the ever-increasing computing resource demands of telecommunications applications is an urgent issue that needs to be addressed in the current technological development of the NFV field. Summary of the Invention

[0005] The embodiments of the present application provide a method, system, and apparatus for managing computing resources, which can meet the ever-increasing computing resource requirements of telecommunication applications in an NFV architecture.

[0006] In a first aspect, an embodiment of the present application provides a communication method applied to a virtualized network function manager (VNFM), the method comprising:

[0007] receiving an instantiation request of a virtualized network function (VNF) from a network function virtualization orchestrator (NFVO), wherein the instantiation request of the VNF includes an operation mode of the VNF;

[0008] Determining addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the VNF and computing resource requirement information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources;

[0009] A computing resource allocation request is sent to the infrastructure manager IM, wherein the computing resource allocation request includes the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF, and the computing resource allocation request is used to request allocation of resources to the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF.

[0010] In the above method, telecommunication applications have different operating modes, and the computing resource requirements under different operating modes are also different. The above method allocates computing resources to VNFs in a heterogeneous computing resource pool according to the operating mode of the VNF, so that the different types of computing resources contained in the heterogeneous computing resource pool can collaboratively execute high-concurrency computing tasks on demand, realize efficient utilization of computing resources, and thus meet the ever-increasing computing resource requirements of telecommunication applications.

[0011] In an optional implementation, the addressing information of the heterogeneous computing resource pool includes identification information of the heterogeneous computing resource pool, and the computing resource requirement information of the VNF includes one or more of the VNF functional framework, the collaborative relationship of computing units in the VNF functional framework, the communication bandwidth requirements between computing units in the VNF functional framework, and the load balancing requirements between computing units in the VNF functional framework.

[0012] In another optional implementation, the VNF operation mode includes a VNF inference mode or a VNF training mode.

[0013] In another optional implementation, the VNF functional framework includes a VNF reasoning framework or a VNF training framework.

[0014] In another optional implementation, the heterogeneous computing resource pool is a computing resource pool stored in the network function virtualization infrastructure NFVI, and the heterogeneous computing resource pool is a resource pool composed of at least two different types of computing resources including a central processing unit CPU, a graphics processing unit GPU, a digital signal processing DSP, an application-specific integrated circuit ASIC, and a field-programmable gate array FPGA.

[0015] In the above method, the heterogeneous computing resources in NFVI are managed by IM in the form of a heterogeneous computing resource pool, which can improve the utilization of heterogeneous computing resources.

[0016] In another optional implementation, the VNF instantiation request further includes a VNF instance identifier, and determining addressing information of a heterogeneous computing resource pool corresponding to the VNF operation mode and computing resource requirement information of the VNF includes:

[0017] Accessing the virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the VNF instantiation request;

[0018] Based on the VNFD, the addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF and the computing resource demand information of the VNF are determined.

[0019] In the above method, the VNFD includes addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF and computing resource demand information of the VNF.

[0020] In yet another optional implementation, determining addressing information of a heterogeneous computing resource pool corresponding to an operating mode of the VNF includes:

[0021] Sending a VNF lifecycle management permission request to the NFVO, wherein the VNF lifecycle management permission request is used to request the addressing information of the heterogeneous computing resource pool corresponding to the VNF operation mode;

[0022] Receive a VNF lifecycle management permission response from the NFVO, where the VNF lifecycle management permission response includes addressing information of the heterogeneous computing resource pool.

[0023] In another optional implementation, the VNF instantiation request further includes a VNF instance identifier, and determining computing resource requirement information of the VNF corresponding to the VNF operation mode includes:

[0024] Accessing the virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the VNF instantiation request;

[0025] Determine the computing resource requirement information of the VNF corresponding to the operation mode of the VNF based on the VNFD.

[0026] In yet another optional implementation, the method further includes:

[0027] receiving a VNF update request from the NFVO, wherein the VNF update request includes an updated operating mode of the VNF;

[0028] Determining addressing information of an updated heterogeneous computing resource pool corresponding to the updated operation mode of the VNF and updated computing resource demand information of the VNF, wherein the updated heterogeneous computing resource pool includes at least two different types of computing resources;

[0029] A computing resource allocation request is sent to the IM, wherein the computing resource allocation request includes the updated addressing information of the heterogeneous computing resource pool and the updated computing resource demand information of the VNF, and the computing resource allocation request is used to request allocation of resources to the VNF in the updated heterogeneous computing resource pool according to the updated addressing information of the heterogeneous computing resource pool and the updated computing resource demand information of the VNF.

[0030] When the VNF's operating mode changes, the above method can re-allocate resources to the VNF from the updated heterogeneous computing resource pool based on the VNF's updated operating mode. Therefore, the above method can flexibly schedule and allocate heterogeneous computing resources to match the computing resource requirements of high-concurrency telecom applications under different operating modes.

[0031] In a second aspect, an embodiment of the present application provides a method for managing computing resources, which is applied to an infrastructure manager IM. The method includes:

[0032] receiving a computing resource allocation request from a virtualized network function manager (VNFM), wherein the computing resource allocation request includes addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the virtualized network function (VNF) and computing resource demand information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources;

[0033] Resources are allocated to VNFs in the heterogeneous computing resource pool based on the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF.

[0034] In the above method, telecommunication applications have different operating modes, and the computing resource requirements under different operating modes are also different. The above method allocates computing resources to VNFs in a heterogeneous computing resource pool according to the operating mode of the VNF, so that the different types of computing resources contained in the heterogeneous computing resource pool can collaboratively execute high-concurrency computing tasks on demand, realize efficient utilization of computing resources, and thus meet the ever-increasing computing resource requirements of telecommunication applications.

[0035] In an optional implementation, after allocating resources to the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource requirement information of the VNF, the method further includes:

[0036] Send a computing resource allocation response to the VNFM, where the computing resource allocation response is used to indicate that resource allocation is completed for the VNF in the heterogeneous computing resource pool.

[0037] In another optional implementation, the addressing information of the heterogeneous computing resource pool includes identification information of the heterogeneous computing resource pool, and the computing resource requirement information of the VNF includes one or more of the VNF functional framework, the collaborative relationship of computing units in the VNF functional framework, the communication bandwidth requirements between computing units in the VNF functional framework, and the load balancing requirements between computing units in the VNF functional framework.

[0038] In another optional implementation, the VNF operation mode includes a VNF inference mode or a VNF training mode.

[0039] In another optional implementation, the VNF functional framework includes a VNF reasoning framework or a VNF training framework.

[0040] In another optional implementation, the heterogeneous computing resource pool is a computing resource pool stored in the network function virtualization infrastructure NFVI, and the heterogeneous computing resource pool is a resource pool composed of at least two different types of computing resources including a central processing unit CPU, a graphics processing unit GPU, a digital signal processing DSP, an application-specific integrated circuit ASIC, and a field-programmable gate array FPGA.

[0041] In the above method, the heterogeneous computing resources in NFVI are managed by IM in the form of a heterogeneous computing resource pool, which can improve the utilization of heterogeneous computing resources.

[0042] In yet another optional implementation, before receiving the computing resource allocation request from the virtualized network function manager VNFM, the method further includes:

[0043] A heterogeneous computing resource pool is created in the network function virtualization infrastructure NFVI according to the computing resource pool description information, wherein the computing resource pool description information includes the computing resource type of the heterogeneous computing resource pool and the topological relationship of the nodes in the heterogeneous computing resource pool.

[0044] In the above method, IM can manage the heterogeneous computing resources stored in NFVI and create a heterogeneous computing resource pool.

[0045] In yet another optional implementation, the topological relationship of the nodes in the heterogeneous computing resource pool is used to reflect the quantity and link connection relationship of at least two different types of computing resources in the heterogeneous computing resource pool.

[0046] In yet another optional implementation, after creating a heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the method further includes:

[0047] Send a computing resource pool management notification, where the computing resource pool management notification includes the addressing information of the heterogeneous computing resource pool and the topological relationship of the nodes in the heterogeneous computing resource pool. The computing resource pool management notification is used to notify the management of the heterogeneous computing resource pool.

[0048] In the above method, after the IM creates the heterogeneous computing resource pool in the NFVI, it can feed back the creation result of the heterogeneous computing resource pool to the upper-level management unit to facilitate the subsequent instantiation of the VNF.

[0049] In yet another optional implementation, before creating the heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the method further includes:

[0050] A computing resource pool creation request is received, wherein the computing resource pool creation request includes computing resource pool description information.

[0051] In the above method, the IM can create multiple heterogeneous computing resource pools in the NFVI based on the computing resource pool creation request from the upper-level management unit.

[0052] In yet another optional implementation, after creating a heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the method further includes:

[0053] Send a computing resource pool creation response, where the computing resource pool creation response is used to indicate that the heterogeneous computing resource pool is successfully created.

[0054] In the above method, IM can create multiple heterogeneous computing resource pools in NFVI based on the computing resource pool creation request from the upper-level management unit, and can also feedback the results of creating the heterogeneous computing resource pools to the upper-level management unit to facilitate the subsequent instantiation of VNFs.

[0055] In yet another optional implementation, the computing resource allocation request further includes addressing information of an updated heterogeneous computing resource pool corresponding to the updated operating mode of the VNF and updated computing resource demand information of the VNF, the updated heterogeneous computing resource pool including at least two different types of computing resources, and the method further includes:

[0056] Resources are allocated to the VNF in the updated heterogeneous computing resource pool according to the updated addressing information of the heterogeneous computing resource pool and the updated computing resource demand information of the VNF.

[0057] When the VNF's operating mode changes, the above method can re-allocate resources to the VNF from the updated heterogeneous computing resource pool based on the VNF's updated operating mode. Therefore, the above method can flexibly schedule and allocate heterogeneous computing resources to match the computing resource requirements of high-concurrency telecom applications under different operating modes.

[0058] In a third aspect, an embodiment of the present application provides a method for managing computing resources, which is applied to a network function virtualization orchestrator (NFVO). The method includes:

[0059] An instantiation request for a virtualized network function VNF is sent to a virtualized network function manager VNFM, wherein the instantiation request for the VNF includes an operation mode of the VNF, and the instantiation request for the VNF is used to request determination of addressing information of a heterogeneous computing resource pool corresponding to the operation mode of the VNF and computing resource demand information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources.

[0060] Receive an instantiation response of the VNF from the VNFM, where the instantiation response of the VNF is used to indicate that the VNF has completed instantiation.

[0061] In another optional implementation, the addressing information of the heterogeneous computing resource pool includes identification information of the heterogeneous computing resource pool, and the computing resource requirement information of the VNF includes one or more of the VNF functional framework, the collaborative relationship of computing units in the VNF functional framework, the communication bandwidth requirements between computing units in the VNF functional framework, and the load balancing requirements between computing units in the VNF functional framework.

[0062] In another optional implementation, the VNF operation mode includes a VNF inference mode or a VNF training mode.

[0063] In another optional implementation, the VNF functional framework includes a VNF reasoning framework or a VNF training framework.

[0064] In another optional implementation, the heterogeneous computing resource pool is a computing resource pool stored in the network function virtualization infrastructure NFVI, and the heterogeneous computing resource pool is a resource pool composed of at least two different types of computing resources including a central processing unit CPU, a graphics processing unit GPU, a digital signal processing DSP, an application-specific integrated circuit ASIC, and a field-programmable gate array FPGA.

[0065] In an optional implementation, before sending the instantiation request of the virtualized network function VNF to the virtualized network function manager VNFM, the method further includes:

[0066] Receive a computing resource pool management notification from the infrastructure manager IM, wherein the computing resource pool management notification is used to indicate the management of a heterogeneous computing resource pool, and the computing resource pool management notification includes addressing information of the heterogeneous computing resource pool and a topological relationship of nodes in the heterogeneous computing resource pool.

[0067] In an optional implementation, after sending the instantiation request of the virtualized network function VNF to the virtualized network function manager VNFM, the method further includes:

[0068] Receive VNF lifecycle management permission request from VNFM;

[0069] Determine the addressing information of the heterogeneous computing resource pool corresponding to the VNF operation mode;

[0070] Send a VNF lifecycle management permission response to the VNFM, where the VNF lifecycle management permission response includes addressing information of the heterogeneous computing resource pool.

[0071] In an optional implementation, before sending the instantiation request of the virtualized network function VNF to the virtualized network function manager VNFM, the method further includes:

[0072] Sending a computing resource pool creation request to the IM, wherein the computing resource pool creation request includes computing resource pool description information, the computing resource pool creation request is used to instruct to create a heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, and the computing resource pool description information includes computing resource types of the heterogeneous computing resource pool and topological relationships of nodes in the heterogeneous computing resource pool;

[0073] In an optional implementation, after sending the computing resource pool creation request to the IM, the method further includes:

[0074] A computing resource pool creation response is received from the IM, wherein the computing resource pool creation response is used to indicate that the heterogeneous computing resource pool is successfully created.

[0075] In an optional implementation, after receiving the VNF instantiation response from the VNFM, the method further includes:

[0076] Send an update request for the VNF to the VNFM, wherein the update request for the VNF includes the updated operating mode of the VNF, and the update request for the VNF is used to request determination of the addressing information of the updated heterogeneous computing resource pool corresponding to the updated operating mode of the VNF and the updated computing resource requirement information of the VNF.

[0077] In a fourth aspect, an embodiment of the present application provides a method for managing computing resources, the method comprising:

[0078] The virtualized network function manager VNFM receives an instantiation request of a virtualized network function VNF, wherein the instantiation request of the VNF includes an operation mode of the VNF; determines addressing information of a heterogeneous computing resource pool corresponding to the operation mode of the VNF and computing resource requirement information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources; and sends a computing resource allocation request to the infrastructure manager IM, wherein the computing resource allocation request includes the addressing information of the heterogeneous computing resource pool and the computing resource requirement information of the VNF;

[0079] The IM receives computing resource allocation requests and allocates resources to the VNF in the heterogeneous computing resource pool based on the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF.

[0080] In an optional implementation, the method further includes:

[0081] The network function virtualization orchestrator NFVO sends an instantiation request of the VNF to the VNFM.

[0082] In an optional implementation, the VNF instantiation request further includes a VNF instance identifier, and the VNFM determines addressing information of a heterogeneous computing resource pool corresponding to the VNF operation mode and computing resource requirement information of the VNF, including:

[0083] The VNFM accesses the VNF virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the VNF instantiation request;

[0084] Based on the VNFD, the VNFM determines the addressing information of the heterogeneous computing resource pool corresponding to the VNF operation mode and the computing resource demand information of the VNF.

[0085] In an optional implementation, the VNFM determines the addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF, including:

[0086] The VNFM sends a VNF lifecycle management permission request to the NFVO, where the VNF lifecycle management permission request is used to request addressing information of the heterogeneous computing resource pool corresponding to the VNF operation mode;

[0087] The NFVO sends a VNF lifecycle management permission response to the VNFM, where the VNF lifecycle management permission response includes addressing information of the heterogeneous computing resource pool.

[0088] In an optional implementation, the VNF instantiation request further includes a VNF instance identifier, and the VNFM determines the computing resource requirement information of the VNF corresponding to the VNF operation mode, including:

[0089] The VNFM accesses the VNF virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the VNF instantiation request;

[0090] The VNFM determines the computing resource requirement information of the VNF corresponding to the operation mode of the VNF based on the VNFD.

[0091] In an optional implementation, after the IM receives the computing resource allocation request and allocates resources to the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource requirement information of the VNF, the method further includes:

[0092] The IM sends a computing resource allocation response to the VNFM, where the computing resource allocation response is used to indicate that resource allocation is completed for the VNF in the heterogeneous computing resource pool;

[0093] The VNFM sends a VNF instantiation response to the NFVO, where the VNF instantiation response is used to indicate that the VNF has completed instantiation.

[0094] In an optional implementation, the method further includes:

[0095] The NFVO sends a VNF update request to the VNFM, where the VNF update request includes the updated operating mode of the VNF;

[0096] The VNFM determines addressing information of an updated heterogeneous computing resource pool corresponding to the updated operation mode of the VNF and updated computing resource demand information of the VNF, wherein the updated heterogeneous computing resource pool includes at least two different types of computing resources;

[0097] The VNFM sends a computing resource allocation request to the IM, where the computing resource allocation request includes updated addressing information of the heterogeneous computing resource pool and updated computing resource requirement information of the VNF;

[0098] The IM allocates resources to the VNF in the updated heterogeneous computing resource pool according to the updated addressing information of the heterogeneous computing resource pool and the updated computing resource demand information of the VNF.

[0099] In a fifth aspect, an embodiment of the present application provides a computing resource management device, which includes a processing module and a transceiver module, and can respectively perform the following operations:

[0100] The transceiver module is used to receive an instantiation request of a virtualized network function (VNF) from a network function virtualization orchestrator (NFVO), wherein the instantiation request of the VNF includes an operating mode of the VNF;

[0101] The processing module is configured to determine addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the VNF and computing resource demand information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources;

[0102] The transceiver module is used to send a computing resource allocation request to the infrastructure manager IM, wherein the computing resource allocation request includes the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF. The computing resource allocation request is used to request the allocation of resources to the VNF in the heterogeneous computing resource pool based on the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF.

[0103] In an optional implementation, the addressing information of the heterogeneous computing resource pool includes identification information of the heterogeneous computing resource pool, and the computing resource requirement information of the VNF includes one or more of the VNF functional framework, the collaborative relationship of computing units in the VNF functional framework, the communication bandwidth requirements between computing units in the VNF functional framework, and the load balancing requirements between computing units in the VNF functional framework.

[0104] In another optional implementation, the VNF operation mode includes a VNF inference mode or a VNF training mode.

[0105] In another optional implementation, the VNF functional framework includes a VNF reasoning framework or a VNF training framework.

[0106] In another optional implementation, the heterogeneous computing resource pool is a computing resource pool stored in the network function virtualization infrastructure NFVI, and the heterogeneous computing resource pool is a resource pool composed of at least two different types of computing resources including a central processing unit CPU, a graphics processing unit GPU, a digital signal processing DSP, an application-specific integrated circuit ASIC, and a field-programmable gate array FPGA.

[0107] In yet another optional implementation, the VNF instantiation request further includes a VNF instance identifier. In determining addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the VNF and computing resource requirement information of the VNF, the processing module is specifically configured to:

[0108] Accessing the virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the VNF instantiation request;

[0109] Based on the VNFD, the addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF and the computing resource demand information of the VNF are determined.

[0110] In yet another optional implementation, in determining the addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF, the processing module is specifically configured to:

[0111] Sending a VNF lifecycle management permission request to the NFVO through the transceiver module, wherein the VNF lifecycle management permission request is used to request addressing information of the heterogeneous computing resource pool corresponding to the VNF operation mode;

[0112] The VNF lifecycle management permission response from the NFVO is received through the transceiver module, wherein the VNF lifecycle management permission response includes addressing information of the heterogeneous computing resource pool.

[0113] In yet another optional implementation, the VNF instantiation request further includes a VNF instance identifier. In determining the computing resource requirement information of the VNF corresponding to the operating mode of the VNF, the processing module is specifically configured to:

[0114] Accessing the virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the VNF instantiation request;

[0115] Determine the computing resource requirement information of the VNF corresponding to the operation mode of the VNF based on the VNFD.

[0116] In yet another optional implementation:

[0117] The transceiver module is further configured to receive a VNF update request from the NFVO, wherein the VNF update request includes an updated operating mode of the VNF;

[0118] The processing module is further configured to determine addressing information of an updated heterogeneous computing resource pool corresponding to the updated operation mode of the VNF and updated computing resource demand information of the VNF, wherein the updated heterogeneous computing resource pool includes at least two different types of computing resources;

[0119] The transceiver module is also used to send a computing resource allocation request to the infrastructure manager IM, wherein the computing resource allocation request includes the updated addressing information of the heterogeneous computing resource pool and the updated computing resource demand information of the VNF, and the computing resource allocation request is used to request allocation of resources to the VNF in the updated heterogeneous computing resource pool according to the updated addressing information of the heterogeneous computing resource pool and the updated computing resource demand information of the VNF.

[0120] In a sixth aspect, an embodiment of the present application provides a computing resource management device, which includes a processing module and a transceiver module, and can respectively perform the following operations:

[0121] The transceiver module is configured to receive a computing resource allocation request from a virtualized network function manager (VNFM), wherein the computing resource allocation request includes addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the virtualized network function (VNF) and computing resource demand information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources;

[0122] The processing module is used to allocate resources to the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF.

[0123] In an optional implementation, after allocating resources to the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource requirement information of the VNF, the transceiver module is further configured to:

[0124] Send a computing resource allocation response to the VNFM, where the computing resource allocation response is used to indicate that resource allocation is completed for the VNF in the heterogeneous computing resource pool.

[0125] In another optional implementation, the addressing information of the heterogeneous computing resource pool includes identification information of the heterogeneous computing resource pool, and the computing resource requirement information of the VNF includes one or more of the VNF functional framework, the collaborative relationship of computing units in the VNF functional framework, the communication bandwidth requirements between computing units in the VNF functional framework, and the load balancing requirements between computing units in the VNF functional framework.

[0126] In another optional implementation, the VNF operation mode includes a VNF inference mode or a VNF training mode.

[0127] In another optional implementation, the VNF functional framework includes a VNF reasoning framework or a VNF training framework.

[0128] In another optional implementation, the heterogeneous computing resource pool is a computing resource pool stored in the network function virtualization infrastructure NFVI, and the heterogeneous computing resource pool is a resource pool composed of at least two different types of computing resources including a central processing unit CPU, a graphics processing unit GPU, a digital signal processing DSP, an application-specific integrated circuit ASIC, and a field-programmable gate array FPGA.

[0129] In yet another optional implementation, before receiving the computing resource allocation request from the virtualized network function manager VNFM, the processing module is further configured to:

[0130] A heterogeneous computing resource pool is created in the network function virtualization infrastructure NFVI according to the computing resource pool description information, wherein the computing resource pool description information includes the computing resource type of the heterogeneous computing resource pool and the topological relationship of the nodes in the heterogeneous computing resource pool.

[0131] In yet another optional implementation, the topological relationship of the nodes in the heterogeneous computing resource pool is used to reflect the quantity and link connection relationship of at least two different types of computing resources in the heterogeneous computing resource pool.

[0132] In yet another optional implementation, after creating a heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the transceiver module is further configured to:

[0133] Send a computing resource pool management notification, where the computing resource pool management notification includes the addressing information of the heterogeneous computing resource pool and the topological relationship of the nodes in the heterogeneous computing resource pool. The computing resource pool management notification is used to notify the management of the heterogeneous computing resource pool.

[0134] In yet another optional implementation, before creating a heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the transceiver module is further configured to:

[0135] A computing resource pool creation request is received, wherein the computing resource pool creation request includes computing resource pool description information.

[0136] In yet another optional implementation, after creating a heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the transceiver module is further configured to:

[0137] Send a computing resource pool creation response, where the computing resource pool creation response is used to indicate that the heterogeneous computing resource pool is successfully created.

[0138] In yet another optional implementation, the computing resource allocation request further includes addressing information of an updated heterogeneous computing resource pool corresponding to the updated operating mode of the VNF and updated computing resource demand information of the VNF, the updated heterogeneous computing resource pool including at least two different types of computing resources, and the processing module is further configured to:

[0139] Resources are allocated to the VNF in the updated heterogeneous computing resource pool according to the updated addressing information of the heterogeneous computing resource pool and the updated computing resource demand information of the VNF.

[0140] In a seventh aspect, an embodiment of the present application provides a computing resource management device, the computing resource management device including a transceiver module, the transceiver module being configured to:

[0141] An instantiation request for a virtualized network function VNF is sent to a virtualized network function manager VNFM, wherein the instantiation request for the VNF includes an operation mode of the VNF, and the instantiation request for the VNF is used to request determination of addressing information of a heterogeneous computing resource pool corresponding to the operation mode of the VNF and computing resource demand information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources.

[0142] Receive an instantiation response of the VNF from the VNFM, where the instantiation response of the VNF is used to indicate that the VNF has completed instantiation.

[0143] In another optional implementation, the addressing information of the heterogeneous computing resource pool includes identification information of the heterogeneous computing resource pool, and the computing resource requirement information of the VNF includes one or more of the VNF functional framework, the collaborative relationship of computing units in the VNF functional framework, the communication bandwidth requirements between computing units in the VNF functional framework, and the load balancing requirements between computing units in the VNF functional framework.

[0144] In another optional implementation, the VNF operation mode includes a VNF inference mode or a VNF training mode.

[0145] In another optional implementation, the VNF functional framework includes a VNF reasoning framework or a VNF training framework.

[0146] In another optional implementation, the heterogeneous computing resource pool is a computing resource pool stored in the network function virtualization infrastructure NFVI, and the heterogeneous computing resource pool is a resource pool composed of at least two different types of computing resources including a central processing unit CPU, a graphics processing unit GPU, a digital signal processing DSP, an application-specific integrated circuit ASIC, and a field-programmable gate array FPGA.

[0147] In an optional implementation, before sending the instantiation request of the virtualized network function VNF to the virtualized network function manager VNFM, the transceiver module is further configured to:

[0148] Receive a computing resource pool management notification from the infrastructure manager IM, wherein the computing resource pool management notification is used to indicate the management of a heterogeneous computing resource pool, and the computing resource pool management notification includes addressing information of the heterogeneous computing resource pool and a topological relationship of nodes in the heterogeneous computing resource pool.

[0149] In an optional implementation, the computing resource management apparatus further includes a processing module, which, after sending a request for instantiation of a virtualized network function VNF to a virtualized network function manager VNFM:

[0150] The transceiver module is also used to receive VNF lifecycle management permission requests from the VNFM;

[0151] The processing module is used to determine the addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF;

[0152] The transceiver module is further used to send a VNF lifecycle management permission response to the VNFM, wherein the VNF lifecycle management permission response includes addressing information of the heterogeneous computing resource pool.

[0153] In an optional implementation, before sending the instantiation request of the virtualized network function VNF to the virtualized network function manager VNFM, the transceiver module is further configured to:

[0154] Sending a computing resource pool creation request to the IM, wherein the computing resource pool creation request includes computing resource pool description information, the computing resource pool creation request is used to instruct to create a heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, and the computing resource pool description information includes computing resource types of the heterogeneous computing resource pool and topological relationships of nodes in the heterogeneous computing resource pool;

[0155] In an optional implementation, after sending the computing resource pool creation request to the IM, the transceiver module is further configured to:

[0156] A computing resource pool creation response is received from the IM, wherein the computing resource pool creation response is used to indicate that the heterogeneous computing resource pool is successfully created.

[0157] In an optional implementation, after receiving the VNF instantiation response from the VNFM, the transceiver module is further configured to:

[0158] Send an update request for the VNF to the VNFM, wherein the update request for the VNF includes the updated operating mode of the VNF, and the update request for the VNF is used to request determination of the addressing information of the updated heterogeneous computing resource pool corresponding to the updated operating mode of the VNF and the updated computing resource requirement information of the VNF.

[0159] In an eighth aspect, an embodiment of the present application provides a computing resource management device, characterized in that it includes a logic circuit and an interface, the logic circuit and the interface are coupled; the interface is used to input and / or output information, wherein:

[0160] The logic circuit is used to execute the method described in the first aspect or any possible implementation of the first aspect, or,

[0161] The logic circuit is used to execute the method described in the second aspect or any possible implementation of the second aspect, or,

[0162] The logic circuit is used to execute the method described in the third aspect or any possible implementation of the third aspect.

[0163] In a ninth aspect, an embodiment of the present application provides a computer-readable storage medium for storing a computer program, wherein:

[0164] When the computer program is executed, it can implement the method of the first aspect or any possible implementation of the first aspect, or,

[0165] When the computer program is executed, it can implement the second aspect or any possible implementation method of the second aspect, or,

[0166] When the computer program is executed, it can implement the method of the third aspect or any possible implementation manner of the third aspect.

[0167] In a tenth aspect, an embodiment of the present application provides a computing resource management system, the communication system including a virtualized network function manager VNFM and an infrastructure manager IM, the VNFM is the device described in the fifth aspect or any possible implementation of the fifth aspect, and the IM is the device described in the sixth aspect or any possible implementation of the sixth aspect.

[0168] In an optional implementation, the communication system further includes a network function virtualization orchestrator NFVO, wherein the NFVO is the device described in the seventh aspect or any possible implementation of the seventh aspect.

[0169] The beneficial effects of the methods, systems and devices provided by any possible implementation of the third aspect and the tenth aspect of this application can refer to the beneficial effects of the technical solutions provided by the first aspect, any possible implementation of the first aspect, the second aspect and any possible implementation of the second aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0170] The following is an introduction to the drawings used in the embodiments of this application.

[0171] FIG1 is a schematic diagram of the architecture of an NFV system provided in an embodiment of the present application;

[0172] FIG2 is a flow chart of a method for managing computing resources provided in an embodiment of the present application;

[0173] FIG3a is a schematic diagram of the distribution of a computing resource pool in an NFVI provided in an embodiment of the present application;

[0174] FIG3 b is a schematic diagram of a topological relationship of nodes in a heterogeneous computing resource pool provided in an embodiment of the present application;

[0175] FIG4 is a flow chart of another method for managing computing resources provided in an embodiment of the present application;

[0176] FIG5 is a flow chart of another method for managing computing resources provided in an embodiment of the present application;

[0177] FIG6 is a flow chart of another method for managing computing resources provided in an embodiment of the present application;

[0178] FIG7 is a schematic diagram of the structure of a computing resource management device provided in an embodiment of the present application;

[0179] FIG8 is a schematic structural diagram of another computing resource management device provided in an embodiment of the present application;

[0180] FIG9 is a schematic structural diagram of another computing resource management device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0181] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.

[0182] Please refer to Figure 1, which is a schematic diagram of the architecture of an NFV system provided in an embodiment of the present application. Specifically, the NFV system includes an NFV management and orchestration system (NFV-MANO) 110, an NFV infrastructure (NFVI) 120, multiple virtualized network functions (VNFs), multiple element management systems (EMSs), and an operations support system / business support system (OSS / BSS) 130. NFV-MANO 110 may include an NFV orchestrator (NFVO) 111, one or more VNF managers (VNFMs), and an infrastructure manager (IM) 112.

[0183] NFVI 120 includes a physical resource layer, which stores various physical resources (also known as hardware resources) required to run VNFs, such as physical computing resources, physical storage resources, and physical network resources. Optionally, NFVI 120 may also include a virtualization layer and a virtual resource layer. The virtualization layer abstracts the physical resources of the physical resource layer, decoupling VNFs from the hardware resource layer and providing VNFs with various virtual resources, such as virtual computing resources (e.g., virtual machines (VMs)), virtual storage resources, and virtual network resources.

[0184] Each VNF is a software implementation of a network function that can run on NFVI 120, different from the physical network function (PNF) in traditional non-virtualized networks. Multiple EMSs correspond one-to-one with multiple VNFs, and each EMS can perform traditional fault, configuration, user, performance, and security management for the VNF.

[0185] NFV-MANO 110 is used to dynamically allocate NFV resources (i.e., various physical resources and / or various virtual resources) within NFVI 120 to support the lifecycle management of multiple VNFs. NFV-MANO 110 can also interoperate with OSS / BSS 130. Specifically, within NFV-MANO 110, NFVO 111 can orchestrate services and resources, manage network services, and integrate VNFs into the NFV architecture. It can also verify and authorize resource requests from the NFV infrastructure. Multiple VNFMs can be responsible for the lifecycle management of VNF instances. IM 112 can be responsible for managing infrastructure resources (including computing resources, storage resources, and network resources) within NFVI 120.

[0186] Currently, when NFV-MANO 110 allocates virtual computing resources to VNF 140 running on NFVI 120, VNFM 113 sends VNF 140's computing resource requirements to NFVO 111 based on the virtual computing resource requirements of VNF 140 described in VNFD, as well as information about requirements for specific physical resources (such as acceleration). Based on this computing resource requirement information, NFVO determines the virtual computing resources required by VNF 140 and, therefore, the target IM identifier to which the computing resources should be allocated. NFVO 111 then returns the target IM identifier to VNFM 113. VNFM 113 sends the computing resource requirement information to IM 112 based on the target IM identifier. The IM then allocates resources to VNF 140 based on the computing resource requirement information within its managed resource range.

[0187] With the rise of new telecom applications with high concurrency (such as 5G video applications, new call applications, AI applications, and computing-network convergence applications), the demand for the processing power of NFVI 120's computing resources to handle these high-concurrency tasks is increasing. Currently, the computing resources in NFVI 120 are primarily homogeneous, represented by CPUs. Because CPUs exhibit significant parallel bottlenecks when handling high-concurrency computing tasks and struggle to overcome latency jitter, relying solely on CPU-based computing resources is insufficient to meet the computing resource requirements of new telecom applications with high concurrency (such as AI analysis and inference).

[0188] Some related technologies deploy heterogeneous computing resources beyond CPUs within NFVI, such as graphics processing units (GPUs), data processing units (DPUs), digital signal processing (DSPs), neural network processing units (NPUs), application-specific integrated circuits (ASICs), and field programmable gate arrays (FPGAs). This forms a heterogeneous computing platform, with the expectation that these heterogeneous computing resources within the platform will meet the computing resource requirements of emerging telecommunications applications through collaborative division of labor. Heterogeneous computing refers to a computing approach that uses computing units with different chip instruction sets and architectures to form a system. CPU+GPU and CPU+FPGA combinations are currently gaining widespread attention. For example, the CPU can handle a wide range of computing tasks and possesses high flexibility and versatility, but its parallel computing capabilities are limited. Graphics processing units (GPUs), on the other hand, have large-scale parallel processing units and high-bandwidth memory, enabling them to execute a large number of computing tasks simultaneously. In other words, the computing capabilities of the CPU and GPU complement each other well.

[0189] The introduction of heterogeneous computing resources into NFVI provides new possibilities for allocating computing resources for telecom applications. However, how to allocate computing resources to better meet the needs of telecom applications (such as high concurrency requirements) has become a pressing issue.

[0190] Please refer to FIG2 , which is a flow chart of a method for managing computing resources provided in an embodiment of the present application. The method can be implemented based on the architecture shown in FIG1 or other architectures. The method includes but is not limited to the following steps:

[0191] Step S201: IM creates multiple heterogeneous computing resource pools in NFVI according to the computing resource pool description information.

[0192] Each heterogeneous computing resource pool in the NFVI includes at least two different types of physical computing resources. For example, a heterogeneous computing resource pool can be composed of at least two different types of physical computing resources, such as CPUs, GPUs, DSPs, NPUs, ASICs, and FPGAs. In practical applications, an IM that manages physical computing resources is also called a physical infrastructure manager (PIM).

[0193] The computing resource description information includes the computing resource type of each heterogeneous computing resource pool and the topological relationship of the nodes (i.e., computing resources) in each heterogeneous computing resource pool. Optionally, the computing resource pool description information may be determined by the IM based on preset computing resource pool planning requirements. The preset computing resource pool planning requirements may be set in the IM by the operator based on the actual computing resource requirements of telecommunications applications.

[0194] Based on description information #1 of a heterogeneous computing resource pool in the computing resource pool description information, the IM determines multiple computing resources in the NFVI that meet the computing resource type specified in description information #1, and organizes the determined multiple computing resources based on the topological relationship specified in description information #1, thereby creating the heterogeneous computing resource pool in the NFVI. It will be understood that the organized computing resources may be all or part of the multiple computing resources determined based on the computing resource type. It will be understood that the creation process of only one heterogeneous computing resource pool in the NFVI is described here as an example; the creation process of other heterogeneous computing resource pools in the NFVI is similar to that of the heterogeneous computing resource pool.

[0195] In an optional embodiment, the multiple heterogeneous computing resource pools created by the IM in the NFVI can be categorized by computing resource type and computing resource specification. For ease of understanding, an exemplary illustration is provided below. Please refer to Figure 3a, which is a schematic diagram of the distribution of computing resource pools in an NFVI provided in an embodiment of the present application. As shown in Figure 3a, NFVI 30 stores a CPU / GPU computing resource pool (specification x) 301, a CPU / GPU computing resource pool (specification y) 302, and a CPU / GPU computing resource pool (specification z) 303. It is understood that the CPU / GPU computing resource pool (specification x) represents a heterogeneous computing resource pool whose computing resource type includes CPUs and GPUs and whose specification is x. The computing resource specification can be understood as a description of the resource characteristics of the computing resource. Resource characteristics can include one or more of resource size, resource type, resource quantity, resource link connection relationship, and vendor information.

[0196] Specifically, the computing resources in each heterogeneous computing resource pool can be understood as a node in the heterogeneous computing resource pool, and each heterogeneous computing resource pool corresponds to a topological relationship of nodes, which is used to reflect the number of computing resources and link connection relationships in the computing resource pool. For example, the topological relationship of the nodes in the CPU / GPU computing resource pool (specification x) 301 can be represented by Figure 3b, which is a schematic diagram of the topological relationship of nodes in a heterogeneous computing resource pool provided in an embodiment of the present application. In a heterogeneous computing resource pool, multiple computing resources of the same type can constitute a computing resource cluster of the corresponding type (such as a CPU cluster or a GPU cluster).

[0197] As shown in Figure 3b, the computing resources of CPU / GPU computing resource pool (specification x) 301 include eight GPUs (i.e., GPU 3010, GPU 3011, ..., GPU 3017) and two CPUs (i.e., CPU 3018, CPU 3019). The eight GPUs can be considered a GPU cluster, and the two CPUs can be considered a CPU cluster.

[0198] Different communication connection modes are represented in the form of different arrows in FIG3b. Exemplarily, as can be seen from FIG3b, each GPU is configured with up to 6 NVIDIA connections (i.e., NVLink) channels, and each GPU is connected to a CPU via a fast peripheral component interconnect (peripheral component interconnect express, PCIe) switch. CPU 3018 is connected to CPU 3019 via a PCIe bus, and each CPU is connected to two external network cards (network interface card, NIC) via a quick path interconnect (quick path interconnect, QPI) bus. Wherein, in order to make 8 GPUs fully connected (i.e., there is a communication channel between any two GPUs), two GPUs that are not connected by the NVLink bus can be connected by a fast peripheral component interconnect (peripheral component interconnect express, PCIe) bus. For example, GPU3010 and GPU3013, as well as GPU3010 and GPU3014, are connected via two NVLink buses. GPU3010 and GPU3011 are connected via a single NVLink bus. However, GPU3010 and GPU3016 cannot be connected via the NVLink bus. Therefore, GPU3010 and GPU3016 are connected via the PCIe bus. Generally speaking, the NVLink bus provides higher bandwidth than the PCIe bus, resulting in better communication performance.

[0199] Optionally, each heterogeneous computing resource pool may also include a storage module instance and an operation and maintenance monitoring instance corresponding to the computing resources.

[0200] It is understandable that the topological relationship of the nodes in the heterogeneous computing resource pool can be combined with the computing capabilities of the computing resources themselves (for example, the number of cores of the computing resources, the core frequency, and the core single clock cycle capability, etc.) and the communication bandwidth between the computing resource combinations, so that the computing resources in the heterogeneous computing resource pool can collaboratively execute computing tasks with high concurrency, thereby meeting the computing resource requirements of new telecommunications applications with high concurrency of computing tasks.

[0201] Optionally, the IM can also create one or more homogeneous computing resource pools (i.e., computing resource pools with the same computing resource type) within the NFVI. As shown in Figure 3a, NFVI 30 can also store CPU computing resource pool (specification a) 304 and CPU computing resource pool (specification b) 305, which can be used to meet the computing resource requirements of telecommunications applications with low computing task concurrency. The process for creating a homogeneous computing resource pool can be the same as that for creating a heterogeneous computing resource pool.

[0202] Step S202: IM sends a computing resource pool management notification to the IM consumer.

[0203] Specifically, after creating multiple heterogeneous computing resource pools in the NFVI, the IM can determine the addressing information for each heterogeneous computing resource pool. The addressing information for the heterogeneous computing resource pool can be its identification information or its name, which can be understood as its identity information. The addressing information can also be used to indicate the computing resource type of the heterogeneous computing resource pool. For example, the name of the heterogeneous computing resource pool can be "CPU / GPU Computing Resource Pool (Specification x)" as shown in Figure 3b, which also indicates that the computing resource types of the heterogeneous computing resource pool include CPUs and GPUs.

[0204] Next, the IM sends a computing resource pool management notification to the IM consumer. In the embodiments of the present application, the IM consumer is a management entity that can invoke functions provided by the IM. For example, the IM consumer can be an NFVO, an OSS / BSS, or a container cluster management (CCM). The computing resource pool management notification includes the addressing information of each heterogeneous computing resource pool in the NFVI, the computing resource type of each heterogeneous computing resource pool, and the topological relationship of the nodes in each heterogeneous computing resource pool. It is understood that the addressing information can explicitly or implicitly include computing resource type information, in which case the notification does not need to include information indicating the computing resource type in addition to the addressing information. For example, the "CPU / GPU computing resource pool (specification x)" can be understood as the addressing information explicitly including computing resource type information, and the specific value range of the addressing information corresponding to a specific computing resource type can be understood as the addressing information implicitly including computing resource type information. Similarly, the topological relationship can also explicitly or implicitly include the computing resource type, in which case the notification does not need to include information indicating the computing resource type in addition to the topological information. It is understandable that this application does not limit the message name of the notification or the message name of the message carrying the notification.

[0205] Step S203: The IM consumer manages multiple heterogeneous computing resource pools.

[0206] Specifically, in response to the notification regarding the management of the computing resource pool, the IM consumer can manage the heterogeneous computing resource pool based on the addressing information of each heterogeneous computing resource pool in the NFVI and store the topological relationship of the nodes in the heterogeneous computing resource pool. In this embodiment of the application, "management" can be understood as obtaining management rights.

[0207] It can be seen that in the embodiment of the present application, IM can organize the physical computing resources in NFVI into multiple heterogeneous computing resource pools based on the preset computing resource pool planning requirements, and feedback the creation results of the heterogeneous computing resource pools in NFVI to the upper-level management unit (i.e., IM consumer) to facilitate the subsequent instantiation of VNF.

[0208] The above describes one method for creating a heterogeneous computing resource pool in NFVI. The following describes another method for creating a heterogeneous computing resource pool in NFVI, in conjunction with Figure 4. Please refer to Figure 4, which is a flowchart of another computing resource management method provided in an embodiment of the present application. This method can be implemented based on the architecture shown in Figure 1 or based on other architectures. The method includes, but is not limited to, the following steps:

[0209] Step S401: the IM consumer sends a computing resource pool creation request to the IM.

[0210] Specifically, the computing resource pool creation request includes computing resource pool description information. For explanations of "computing resource pool description information" and "IM consumer", please refer to the corresponding descriptions of steps S201 and S203 in the embodiment shown in Figure 2, and will not be repeated here.

[0211] Optionally, the computing resource pool creation request may further include addressing information of the heterogeneous computing resource pool corresponding to the computing resource pool description information.

[0212] Step S402: IM creates multiple heterogeneous computing resource pools in NFVI according to the computing resource pool description information.

[0213] S402 may specifically refer to the process of creating multiple heterogeneous computing resource pools in S201 above. The difference is that the heterogeneous computing resource pool created in S402 may include virtual computing resources. Each heterogeneous computing resource pool includes at least two different types of virtual computing resources. For example, a heterogeneous computing resource pool may be a resource pool composed of at least two different types of virtual computing resources among virtual CPU (also called vCPU), virtual GPU (also called vGPU), virtual DSP (also called vDSP), virtual NPU (also called vNPU), virtual ASIC (also called vASIC) and virtual FPGA (also called vFPGA). In actual applications, the IM that can manage virtual computing resources may also be called a virtual infrastructure manager (VIM). Specifically, in response to the computing resource pool creation request, the IM virtualizes multiple physical computing resources in the NFVI to form multiple virtual computing resources in the NFVI.

[0214] After creating multiple heterogeneous computing resource pools in the NFVI, the IM can store the addressing information for each of the heterogeneous computing resource pools and the topological relationships of the nodes within each heterogeneous computing resource pool. For explanations of the "addressing information of the heterogeneous computing resource pools" and "topological relationships of the nodes within the heterogeneous computing resource pools," refer to the corresponding descriptions of step S202 and step S201 in the embodiment shown in Figure 2 , and are not repeated here.

[0215] Step S403: The IM sends a computing resource pool creation response to the IM consumer.

[0216] Specifically, after successfully creating multiple heterogeneous computing resource pools in NFVI, IM sends a computing resource pool creation response to the IM consumer. The computing resource pool creation response is used to indicate that the multiple heterogeneous computing resource pools are successfully created.

[0217] Optionally, if the IM fails to successfully create multiple heterogeneous computing resource pools in the NFVI (i.e., a heterogeneous computing resource pool creation failure occurs), the IM may also send a computing resource pool creation response to the IM consumer. In this case, the computing resource pool creation response may be used to indicate that a heterogeneous computing resource pool creation failure occurred. Specifically, the computing resource pool creation response may include the identifier of the heterogeneous computing resource pool that failed to be created, and optionally, the reason for the creation failure. For example, the reason for the creation failure may be that there are no computing resources in the NFVI that match the computing resource type of a heterogeneous computing resource pool in the computing resource pool description information.

[0218] It can be seen that in the embodiment of the present application, IM can virtualize the physical computing resources in NFVI into virtual computing resources and then organize them into multiple heterogeneous computing resource pools based on the computing resource pool creation request from the upper-level management unit (i.e., IM consumer), and feedback the results of creating the heterogeneous computing resource pool to the upper-level management unit to facilitate the subsequent instantiation of VNF.

[0219] Compared to the embodiment shown in Figure 2, the process for creating heterogeneous computing resource pools in the embodiment shown in Figure 4 is more flexible. The number and type of heterogeneous computing resource pools in the NFVI can be adjusted based on the computing resource pool creation request. Furthermore, because the computing resources in the heterogeneous computing resource pools in the embodiment shown in Figure 4 are virtual computing resources formed by virtualizing physical computing resources, the utilization rate of physical computing resources during the creation of heterogeneous computing resource pools in the embodiment shown in Figure 4 is higher than that in the embodiment shown in Figure 2.

[0220] The following describes the VNF instantiation process in conjunction with Figure 5. Please refer to Figure 5, which is a flowchart of another computing resource management method provided in an embodiment of the present application. This method can be implemented based on the architecture shown in Figure 1 or other architectures. The method includes but is not limited to the following steps:

[0221] Step S501: NFVO sends a VNF instantiation request to VNFM.

[0222] Specifically, the VNF is used to implement telecommunication applications (such as 5G video applications, new call applications, AI applications, and computing-network integration applications). For example, the VNF may be VNF 140 shown in FIG1 .

[0223] The instantiation request of the VNF includes the operating mode of the VNF. The operating mode of the VNF can be the operating mode when the VNF performs computing tasks to implement analysis or reasoning functions. Optionally, the operating mode of the VNF can be a VNF reasoning mode or a VNF training mode. The VNF reasoning mode can be further classified into a computing-intensive reasoning mode and a data-intensive reasoning mode, etc. The VNF training mode can be further classified into a small-model-based training mode and a large-model-based training mode, etc., and the VNF training mode can be attached with information about the training platform manufacturer. It is understandable that different VNF ​​operating modes correspond to different computing resource requirements and computational complexity.

[0224] It is understood that the VNF operating mode carried in the instantiation request is used to search for heterogeneous or homogeneous computing resources suitable for the VNF's operating mode. Other VNF description information that corresponds to computing resource requirements and / or computational complexity can replace the VNF operating mode here to search for heterogeneous or homogeneous computing resources suitable for the VNF's description information.

[0225] The VNF instantiation request may also include a VNF instance identifier, which may be understood as information that uniquely identifies the VNF instance.

[0226] Step S502: The VNFM determines the addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF and the computing resource demand information of the VNF.

[0227] Specifically, the heterogeneous computing resource pool corresponding to the operating mode of the VNF is a heterogeneous computing resource pool stored in multiple heterogeneous computing resource pools in the NFVI and serves as a carrier for the operation of the VNF. The present application does not limit whether the computing resources in the heterogeneous computing resource pool are virtual computing resources or physical computing resources. The present application does not limit the creation process of multiple heterogeneous computing resource pools in the NFVI, and the methods described in Figures 2 and 4 are only examples. For the explanation of the "addressing information of the heterogeneous computing resource pool", please refer to the description of the corresponding part in step S202 in the embodiment shown in Figure 2, and will not be repeated here. For the convenience of description, the "heterogeneous computing resource pool corresponding to the operating mode of the VNF" will be referred to as the "target heterogeneous computing resource pool" below.

[0228] The computing resource requirement information of the VNF corresponding to the operation mode of the VNF includes one or more of the VNF functional framework, the collaborative relationship of the computing units in the VNF functional framework, and the communication requirements between the computing units in the VNF functional framework. The communication requirements between the computing units may be bandwidth requirements and / or load balancing requirements. Optionally, corresponding to the operation mode of the VNF, the VNF functional framework may be a VNF reasoning framework or a VNF training framework. The VNF reasoning framework can be further classified into a computationally intensive reasoning framework and a data-intensive reasoning framework, etc., and the VNF training framework can be further classified into a small-model-based training framework and a large-model-based training framework, etc. For ease of description, the "computing resource requirement information of the VNF corresponding to the operation mode of the VNF" will be referred to as "target computing resource requirement information" below.

[0229] In response to the VNF instantiation request, the VNFM accesses the VNF's virtualized network function descriptor (VNFD) file based on the VNF instance identifier. The VNFD includes information reflecting the correspondence between the VNF's operating mode and target computing resource requirements. The VNFM can then determine the target computing resource requirements based on the VNFD. The target computing resource requirements can be defined in the VNFD in the form of an intent.

[0230] In an optional embodiment, the VNFD further includes information reflecting the correspondence between the operating mode of the VNF and the addressing information of the target heterogeneous computing resource pool. The VNFM can determine the addressing information of the target heterogeneous computing resource pool based on the VNFD.

[0231] The VNFD in this embodiment includes addressing information for heterogeneous computing resource pools. It is understood that this embodiment can be applied to scenarios where multiple heterogeneous computing resource pools are planned and deployed in advance within the NFVI. In other words, the process for creating multiple heterogeneous computing resource pools within the NFVI in this embodiment can be the method described in Figure 2.

[0232] When the creation process for multiple heterogeneous computing resource pools in the NFVI follows the method described in FIG2 , the information in the VNFD reflecting the correspondence between the VNF's operating mode and the addressing information of the target heterogeneous computing resource pool can be determined based on the addressing information, computing resource types, and topological relationships of the heterogeneous computing resource pools in the NFVI. In other words, the VNFM can determine, based on the acquired heterogeneous computing resource pool information (including the aforementioned addressing information, computing resource type information, and topological relationships), a suitable heterogeneous resource pool for each of the one or more operating modes of the VNF instance, and configure the correspondence between the operating mode and the addressing information of the heterogeneous resource pools in the VNFD. The VNFM can obtain information about the heterogeneous computing resource pools through direct or indirect notifications of computing resource pool management. It is understood that, as previously described, when the addressing information or topological relationships explicitly or implicitly include computing resource type information, the correspondence between the VNF's operating mode and the addressing information of the target heterogeneous computing resource pool in the VNFD can be determined directly based on the addressing information and topological relationships of the heterogeneous resource pools.

[0233] It should be noted that the setting of the information in the VNFD that reflects the correspondence between the operating mode of the VNF and the addressing information of the target heterogeneous computing resource pool can be completed autonomously by the NFV system or manually.

[0234] In another optional embodiment, after determining the target computing resource requirement information, the VNFM sends a VNF lifecycle management permission request to the NFVO, and the VNF lifecycle management permission request includes the operating mode information of the VNF. Optionally, the VNF lifecycle management permission request may also include the target computing resource requirement information. Optionally, the operating mode information included in the VNF lifecycle management permission request may be the operating mode of the VNF included in the VNF instantiation request in step S501, or may be the operating mode determined by the VNFM after further refining the operating mode of the VNF based on the target computing resource requirement information. For example, if the operating mode of the VNF is the reasoning mode, the operating mode information carried in the VNF lifecycle management permission request may be a data-intensive application reasoning mode. In response to the VNF lifecycle management permission request, the NFVO determines the addressing information of the target heterogeneous computing resource pool based on the operating mode information, and sends a VNF lifecycle management permission response to the VNFM. Optionally, when the VNF lifecycle management permission request includes target computing resource demand information, the NFVO can select heterogeneous computing resources that meet the target computing resource demand when determining the addressing information of the target heterogeneous computing resource pool, and determine the corresponding addressing information. The VNF lifecycle management permission response includes the addressing information of the target heterogeneous computing resource pool. The VNFM can determine the addressing information of the target heterogeneous computing resource pool based on the VNF lifecycle management permission response. It should be noted that "VNF lifecycle management permission request" and "VNF lifecycle management permission response" are exemplary names for the two messages in the embodiments of the present application. The two messages can also be named by other names, which are not limited here.

[0235] In this implementation, the addressing information for the heterogeneous computing resource pool is determined by the NFVO and then sent to the VNFM. It is understood that this implementation can also be applied to situations where the process for creating multiple heterogeneous computing resource pools in the NFVI is relatively flexible. In other words, the process for creating multiple heterogeneous computing resource pools in the NFVI in this implementation can be the method described in Figure 4 , in which case the IM consumer is the NFVO.

[0236] Step S503: The VNFM sends a computing resource allocation request to the IM.

[0237] Specifically, the computing resource allocation request includes addressing information of a target heterogeneous computing resource pool and target computing resource demand information.

[0238] Step S504: The IM allocates resources to the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF.

[0239] Specifically, in response to the computing resource allocation request, the IM determines the target heterogeneous computing resource pool from multiple heterogeneous computing resource pools in the NFVI according to the addressing information of the target heterogeneous computing resource pool, and allocates resources to the VNF in the target heterogeneous computing resource pool according to the target computing resource demand information.

[0240] Optionally, the target computing resource requirement information can be described in the form of intent. The IM can parse the target computing resource requirement described in the form of intent through the built-in knowledge base, determine the combination of computing resources required under the operating mode of the VNF in the target heterogeneous computing resource pool, and allocate the determined computing resources to the VNF.

[0241] Optionally, only one heterogeneous resource pool may be created in S201 or S402 above. Then, in S504, the IM determines to allocate resources to the VNF in the heterogeneous resource pool in the NFVI according to the above addressing information.

[0242] This embodiment may further include the following steps:

[0243] Step S505: The IM sends a computing resource allocation response to the VNFM.

[0244] Specifically, after completing resource allocation for the VNF, the IM sends a computing resource allocation response to the VNFM. The computing resource allocation response is used to indicate that resource allocation is completed for the VNF in the target heterogeneous computing resource pool.

[0245] Step S506: The VNFM sends a VNF instantiation response to the NFVO.

[0246] Specifically, in response to the computing resource allocation response, the VNFM sends an instantiation response of the VNF to the NFVO, where the instantiation response of the VNF is used to indicate that the VNF has completed instantiation.

[0247] It can be seen from this that the embodiment of the present application is based on the heterogeneous computing resource pool in the NFVI, and allocates computing resources to the VNF in the heterogeneous computing resource pool according to the operating mode of the VNF, so that the different types of computing resources contained in the heterogeneous computing resource pool can collaboratively execute high-concurrency computing tasks on demand, thereby achieving efficient utilization of computing resources. Based on the embodiment of the present application, it is possible to guide telecommunication applications to the corresponding heterogeneous computing resource pool on demand for lifecycle management. Therefore, the embodiment of the present application can not only improve the utilization rate of heterogeneous computing resources in the NFVI, but also meet the ever-increasing computing resource requirements of telecommunication applications, so that the computing resource requirements of new telecommunication applications with high concurrency can also be met.

[0248] Furthermore, when the operating mode of an instantiated VNF changes, the VNF can be re-instantiated. Please refer to Figure 6, which is a flow chart of another method for managing computing resources provided by an embodiment of the present application. This method can be implemented based on the architecture shown in Figure 1 or other architectures, and includes but is not limited to the following steps:

[0249] Step S601: NFVO sends a VNF update request to VNFM.

[0250] Specifically, the VNF may be the VNF in the embodiment shown in FIG5 . The VNF update request includes the updated operating mode of the VNF and the instantiation identifier of the VNF. For example, the initial operating mode of the VNF may be the VNF inference mode, and the updated operating mode of the VNF may be the VNF training mode. For explanations of the "VNF operating mode" and the "VNF instantiation identifier," please refer to the corresponding description of step S501 in the embodiment shown in FIG5 , which will not be repeated here.

[0251] Step S602: The VNFM determines the updated addressing information of the heterogeneous computing resource pool corresponding to the updated operation mode of the VNF and the updated computing resource demand information of the VNF.

[0252] It is understood that when the VNF's operating mode changes, the heterogeneous computing resource pool corresponding to the VNF's operating mode and the VNF's computing resource demand information will also change accordingly. The specific implementation of step S602 is explained in the embodiment shown in Figure 5, and will not be repeated here.

[0253] Step S603: The VNFM sends a computing resource allocation request to the IM.

[0254] For the specific implementation of step S603, please refer to the explanation of step S503 in the embodiment shown in FIG5 , which will not be repeated here.

[0255] Step S604: The IM allocates resources to the VNF in the updated heterogeneous computing resource pool according to the updated addressing information of the heterogeneous computing resource pool and the updated computing resource demand information of the VNF.

[0256] For the specific implementation of step S604, please refer to the explanation of step S504 in the embodiment shown in FIG5 , which will not be repeated here.

[0257] Step S605: The IM sends a computing resource allocation response to the VNFM.

[0258] For the specific implementation of step S605, please refer to the explanation of step S505 in the embodiment shown in FIG5 , which will not be repeated here.

[0259] Step S606: The VNFM sends a VNF instantiation response to the NFVO.

[0260] For the specific implementation of step S606, please refer to the explanation of step S506 in the embodiment shown in FIG5 , which will not be repeated here.

[0261] Thus, in the case of a change in the VNF's operating mode, the embodiment of the present application re-allocates resources to the VNF in the updated heterogeneous computing resource pool according to the VNF's updated operating mode. Therefore, the embodiment of the present application can flexibly schedule heterogeneous computing resources in the NFVI to match the computing resource requirements of telecommunications applications with high computing task concurrency under different operating modes.

[0262] In addition, the computing resource management method provided in the embodiments of the present application can serve as a technical reference for the management method of network resources and / or storage resources in the NFV architecture. For example, referring to the method described in FIG5 , the IM may create a heterogeneous network resource pool and / or a heterogeneous storage resource pool in the NFVI in the same manner as creating a heterogeneous computing resource pool, and allocate resources to the VNFs in the heterogeneous network resource pool and / or heterogeneous storage resource pool as needed by interacting with the VNFM and NFVO, thereby meeting the storage requirements (such as storage acceleration) and / or network requirements (such as network acceleration) of new telecommunications applications.

[0263] The following will introduce the computing resource management device provided in the embodiments of the present application.

[0264] The present application divides the computing resource management device into functional modules according to the above-mentioned method embodiment. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in this application is schematic and is only a logical functional division. There may be other division methods in actual implementation. The computing resource management device of the embodiment of the present application will be described in detail below with reference to Figures 7 to 9.

[0265] FIG7 is a schematic diagram of the structure of a computing resource management device provided in an embodiment of the present application. As shown in FIG7 , the computing resource management device 70 includes a processing module 701 and a transceiver module 702. The transceiver module 702 can implement corresponding communication functions, and the processing module 701 is used for data processing. For example, the transceiver module 702 can also be referred to as an interface, a communication interface, or a communication module.

[0266] In some embodiments of the present application, the computing resource management device 70 can be used to perform the actions performed by the VNFM side in the above method embodiments. For example, the computing resource management device 70 can be the VNFM itself or a chip or functional module that can be configured in the VNFM. The transceiver module 702 is used to perform the VNFM transceiver-related operations in the above method embodiments, and the processing module 701 is used to perform the VNFM processing-related operations in the above method embodiments. The processing module 701 can perform the corresponding operations by calling a computer program or by performing the corresponding operations through corresponding hardware circuits. The transceiver module 702 can perform the transceiver operations independently or under the control of the processing module 701.

[0267] Exemplarily, the computing resource management device 70 shown in FIG7 may be a VNFM or a device in a VNFM. The processing module 701 and the transceiver module 702 in the computing resource management device 70 may respectively perform the following operations:

[0268] The transceiver module 702 is configured to receive an instantiation request for a virtualized network function (VNF) from a network function virtualization orchestrator (NFVO), wherein the instantiation request for the VNF includes an operating mode of the VNF;

[0269] The processing module 701 is configured to determine addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the VNF and computing resource requirement information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources;

[0270] The transceiver module 702 is used to send a computing resource allocation request to the infrastructure manager IM, wherein the computing resource allocation request includes the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF, and the computing resource allocation request is used to request the allocation of resources to the VNF in the heterogeneous computing resource pool based on the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF.

[0271] In an optional implementation, the addressing information of the heterogeneous computing resource pool includes identification information of the heterogeneous computing resource pool, and the computing resource requirement information of the VNF includes one or more of the VNF functional framework, the collaborative relationship of computing units in the VNF functional framework, the communication bandwidth requirements between computing units in the VNF functional framework, and the load balancing requirements between computing units in the VNF functional framework.

[0272] In another optional implementation, the VNF operation mode includes a VNF inference mode or a VNF training mode.

[0273] In another optional implementation, the VNF functional framework includes a VNF reasoning framework or a VNF training framework.

[0274] In another optional implementation, the heterogeneous computing resource pool is a computing resource pool stored in the network function virtualization infrastructure NFVI, and the heterogeneous computing resource pool is a resource pool composed of at least two different types of computing resources including a central processing unit CPU, a graphics processing unit GPU, a digital signal processing DSP, an application-specific integrated circuit ASIC, and a field-programmable gate array FPGA.

[0275] In another optional implementation, the VNF instantiation request further includes a VNF instance identifier. In determining the addressing information of the heterogeneous computing resource pool corresponding to the VNF operation mode and the computing resource requirement information of the VNF, the processing module 701 is specifically configured to:

[0276] Accessing the virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the VNF instantiation request;

[0277] Based on the VNFD, the addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF and the computing resource demand information of the VNF are determined.

[0278] In yet another optional implementation, in determining the addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF, the processing module 701 is specifically configured to:

[0279] Sending a VNF lifecycle management permission request to the NFVO through the transceiver module 702, wherein the VNF lifecycle management permission request is used to request to determine the addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF;

[0280] The VNF lifecycle management permission response from the NFVO is received through the transceiver module 702, wherein the VNF lifecycle management permission response includes addressing information of the heterogeneous computing resource pool.

[0281] In another optional implementation, the VNF instantiation request further includes a VNF instance identifier. In determining the computing resource requirement information of the VNF corresponding to the operation mode of the VNF, the processing module 701 is specifically configured to:

[0282] Accessing the virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the VNF instantiation request;

[0283] Determine the computing resource requirement information of the VNF corresponding to the operation mode of the VNF based on the VNFD.

[0284] In yet another optional implementation:

[0285] The transceiver module 702 is further configured to receive a VNF update request from the NFVO, wherein the VNF update request includes an updated operating mode of the VNF;

[0286] The processing module 701 is further configured to determine addressing information of an updated heterogeneous computing resource pool corresponding to the updated operation mode of the VNF and updated computing resource demand information of the VNF, wherein the updated heterogeneous computing resource pool includes at least two different types of computing resources;

[0287] The transceiver module 702 is also used to send a computing resource allocation request to the infrastructure manager IM, wherein the computing resource allocation request includes the updated addressing information of the heterogeneous computing resource pool and the updated computing resource demand information of the VNF, and the computing resource allocation request is used to request allocation of resources to the VNF in the updated heterogeneous computing resource pool according to the updated addressing information of the heterogeneous computing resource pool and the updated computing resource demand information of the VNF.

[0288] Referring to Figure 7 , in other embodiments of the present application, the computing resource management device 70 can be used to execute the actions performed on the IM side in the above method embodiments. For example, the computing resource management device 70 can be the IM itself or a chip or functional module configured in the IM. The transceiver module 702 is used to execute the operations related to IM transceiver in the above method embodiments, and the processing module 701 is used to execute the operations related to IM processing in the above method embodiments. The processing module 701 can execute the corresponding operations by calling a computer program or by executing the corresponding hardware circuit. The transceiver module 702 can execute the corresponding transceiver operations independently or under the control of the processing module 701.

[0289] For example, the computing resource management device 70 shown in FIG7 may be an IM or a device in the IM. The processing module 701 and the transceiver module 702 in the computing resource management device 70 may respectively perform the following operations:

[0290] The transceiver module 702 is configured to receive a computing resource allocation request from the virtualized network function manager VNFM, wherein the computing resource allocation request includes addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the virtualized network function VNF and computing resource demand information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources;

[0291] The processing module 701 is used to allocate resources to the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF.

[0292] In an optional implementation, after allocating resources to the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource requirement information of the VNF, the transceiver module 702 is further configured to:

[0293] Send a computing resource allocation response to the VNFM, where the computing resource allocation response is used to indicate that resource allocation is completed for the VNF in the heterogeneous computing resource pool.

[0294] In another optional implementation, the addressing information of the heterogeneous computing resource pool includes identification information of the heterogeneous computing resource pool, and the computing resource requirement information of the VNF includes one or more of the VNF functional framework, the collaborative relationship of computing units in the VNF functional framework, the communication bandwidth requirements between computing units in the VNF functional framework, and the load balancing requirements between computing units in the VNF functional framework.

[0295] In another optional implementation, the VNF operation mode includes a VNF inference mode or a VNF training mode.

[0296] In another optional implementation, the VNF functional framework includes a VNF reasoning framework or a VNF training framework.

[0297] In another optional implementation, the heterogeneous computing resource pool is a computing resource pool stored in the network function virtualization infrastructure NFVI, and the heterogeneous computing resource pool is a resource pool composed of at least two different types of computing resources including a central processing unit CPU, a graphics processing unit GPU, a digital signal processing DSP, an application-specific integrated circuit ASIC, and a field-programmable gate array FPGA.

[0298] In yet another optional implementation, before receiving the computing resource allocation request from the virtualized network function manager VNFM, the processing module 701 is further configured to:

[0299] A heterogeneous computing resource pool is created in the network function virtualization infrastructure NFVI according to the computing resource pool description information, wherein the computing resource pool description information includes the computing resource type of the heterogeneous computing resource pool and the topological relationship of the nodes in the heterogeneous computing resource pool.

[0300] In yet another optional implementation, the topological relationship of the nodes in the heterogeneous computing resource pool is used to reflect the quantity and link connection relationship of at least two different types of computing resources in the heterogeneous computing resource pool.

[0301] In yet another optional implementation, after creating a heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the transceiver module 702 is further configured to:

[0302] Send a computing resource pool management notification, where the computing resource pool management notification includes the addressing information of the heterogeneous computing resource pool and the topological relationship of the nodes in the heterogeneous computing resource pool. The computing resource pool management notification is used to notify the management of the heterogeneous computing resource pool.

[0303] In yet another optional implementation, before creating a heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the transceiver module 702 is further configured to:

[0304] A computing resource pool creation request is received, wherein the computing resource pool creation request includes computing resource pool description information.

[0305] In yet another optional implementation, after creating a heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the transceiver module 702 is further configured to:

[0306] Send a computing resource pool creation response, where the computing resource pool creation response is used to indicate that the heterogeneous computing resource pool is successfully created.

[0307] In another optional implementation, the computing resource allocation request further includes addressing information of an updated heterogeneous computing resource pool corresponding to the updated operation mode of the VNF and updated computing resource demand information of the VNF, the updated heterogeneous computing resource pool includes at least two different types of computing resources, and the processing module 701 is further configured to:

[0308] Resources are allocated to the VNF in the updated heterogeneous computing resource pool according to the updated addressing information of the heterogeneous computing resource pool and the updated computing resource demand information of the VNF.

[0309] Referring to Figure 7 , in other embodiments of the present application, the computing resource management device 70 can be used to execute the actions performed on the NFVO side in the above method embodiments. For example, the computing resource management device 70 can be the NFVO itself or a chip or functional module configured within the NFVO. The transceiver module 702 is used to execute the NFVO transceiver-related operations in the above method embodiments, and the processing module 701 is used to execute the NFVO processing-related operations in the above method embodiments. The processing module 701 can execute the corresponding operations by calling a computer program or by executing the corresponding hardware circuits. The transceiver module 702 can execute the corresponding transceiver operations independently or under the control of the processing module 701.

[0310] Exemplarily, the computing resource management device 70 shown in FIG7 may be an NFVO or a device in the NFVO. The transceiver module 702 in the computing resource management device 70 is configured to:

[0311] An instantiation request for a virtualized network function VNF is sent to a virtualized network function manager VNFM, wherein the instantiation request for the VNF includes an operation mode of the VNF, and the instantiation request for the VNF is used to request determination of addressing information of a heterogeneous computing resource pool corresponding to the operation mode of the VNF and computing resource demand information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources.

[0312] Receive an instantiation response of the VNF from the VNFM, where the instantiation response of the VNF is used to indicate that the VNF has completed instantiation.

[0313] In another optional implementation, the addressing information of the heterogeneous computing resource pool includes identification information of the heterogeneous computing resource pool, and the computing resource requirement information of the VNF includes one or more of the VNF functional framework, the collaborative relationship of computing units in the VNF functional framework, the communication bandwidth requirements between computing units in the VNF functional framework, and the load balancing requirements between computing units in the VNF functional framework.

[0314] In another optional implementation, the VNF operation mode includes a VNF inference mode or a VNF training mode.

[0315] In another optional implementation, the VNF functional framework includes a VNF reasoning framework or a VNF training framework.

[0316] In another optional implementation, the heterogeneous computing resource pool is a computing resource pool stored in the network function virtualization infrastructure NFVI, and the heterogeneous computing resource pool is a resource pool composed of at least two different types of computing resources including a central processing unit CPU, a graphics processing unit GPU, a digital signal processing DSP, an application-specific integrated circuit ASIC, and a field-programmable gate array FPGA.

[0317] In an optional implementation, before sending the instantiation request of the virtualized network function VNF to the virtualized network function manager VNFM, the transceiver module 702 is further configured to:

[0318] Receive a computing resource pool management notification from the infrastructure manager IM, wherein the computing resource pool management notification is used to indicate the management of a heterogeneous computing resource pool, and the computing resource pool management notification includes addressing information of the heterogeneous computing resource pool and a topological relationship of nodes in the heterogeneous computing resource pool.

[0319] In an optional implementation, after sending a request for instantiation of a virtualized network function VNF to a virtualized network function manager VNFM:

[0320] The transceiver module 702 is further configured to receive a VNF lifecycle management permission request from the VNFM;

[0321] The processing module 701 is used to determine the addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF;

[0322] The transceiver module 702 is further used to send a VNF lifecycle management permission response to the VNFM, wherein the VNF lifecycle management permission response includes addressing information of the heterogeneous computing resource pool.

[0323] In an optional implementation, before sending the instantiation request of the virtualized network function VNF to the virtualized network function manager VNFM, the transceiver module 702 is further configured to:

[0324] Sending a computing resource pool creation request to the IM, wherein the computing resource pool creation request includes computing resource pool description information, the computing resource pool creation request is used to instruct to create a heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, and the computing resource pool description information includes computing resource types of the heterogeneous computing resource pool and topological relationships of nodes in the heterogeneous computing resource pool;

[0325] In an optional implementation, after sending the computing resource pool creation request to the IM, the transceiver module 702 is further configured to:

[0326] A computing resource pool creation response is received from the IM, wherein the computing resource pool creation response is used to indicate that the heterogeneous computing resource pool is successfully created.

[0327] In an optional implementation, after receiving the VNF instantiation response from the VNFM, the transceiver module 702 is further configured to:

[0328] Send an update request for the VNF to the VNFM, wherein the update request for the VNF includes the updated operating mode of the VNF, and the update request for the VNF is used to request determination of the addressing information of the updated heterogeneous computing resource pool corresponding to the updated operating mode of the VNF and the updated computing resource requirement information of the VNF.

[0329] The specific descriptions of the transceiver module 702 and the processing module 701 shown in the above embodiments are only examples. For the specific functions or execution steps of the transceiver module 702 and the processing module 701, please refer to the above method embodiments and will not be described in detail here.

[0330] The above describes the computing resource management device of the embodiment of the present application. The following describes possible product forms of the computing resource management device. Any product having the functions of the computing resource management device described in FIG7 falls within the scope of protection of the embodiment of the present application.

[0331] The following introduction is for illustrative purposes only and does not limit the product form of the computing resource management device of the embodiment of the present application to this.

[0332] In one possible implementation, in the computing resource management device 70 shown in FIG7 , the processing module 701 may be one or more processors, the transceiver module 702 may be a transceiver, or the transceiver module 702 may be a sending module and a receiving module, the sending module may be a transmitter, the receiving module may be a receiver, and the sending module and the receiving module are integrated into a single device, such as a transceiver. In the embodiment of the present application, the processor and the transceiver may be coupled, etc., and the embodiment of the present application does not limit the connection method between the processor and the transceiver. During the execution of the above method, the process of sending information in the above method may be the process of the processor outputting the above information. When outputting the above information, the processor outputs the above information to the transceiver so that the transceiver transmits it. After being output by the processor, the above information may also need to undergo other processing before reaching the transceiver. Similarly, the process of receiving information in the above method may be the process of the processor receiving the input information. When the processor receives the input information, the transceiver receives the above information and inputs it into the processor. Furthermore, after the transceiver receives the above information, the above information may need to be processed further before being input into the processor.

[0333] As shown in Figure 8 , the computing resource management device 80 includes one or more processors 802 and a transceiver 801. For example, the transceiver 801 is configured to execute the functions or steps implemented by the transceiver module 702 shown in Figure 7 , and the processor 802 is configured to execute the functions or steps implemented by the processing module 701 shown in Figure 7 . For detailed descriptions of the processor 802 and the transceiver 801, please refer to Figure 7 or the method embodiments shown above and will not be described in detail here.

[0334] In the above-mentioned embodiments, the description of the relevant steps and information can be referred to the introduction in the above method embodiment, and will not be described in detail here.

[0335] In various implementations of the computing resource management apparatus shown in FIG8 , the transceiver may include a receiver and a transmitter, wherein the receiver is configured to perform a receiving function (or operation) and the transmitter is configured to perform a transmitting function (or operation). The transceiver is configured to communicate with other devices / apparatuses via a transmission medium.

[0336] Optionally, the computing resource management device 80 may further include one or more memories 803 for storing program instructions and / or data. The memory 803 is coupled to the processor 802. The coupling in the embodiment of the present application is an indirect coupling or communication connection between devices, units or modules, which can be electrical, mechanical or other forms, and is used for information exchange between devices, units or modules. The processor 802 may operate in conjunction with the memory 803. The processor 802 may execute program instructions stored in the memory 803. Optionally, at least one of the above-mentioned one or more memories may be included in the processor.

[0337] The specific connection medium between the transceiver 801, processor 802, and memory 803 is not limited in the embodiments of the present application. In FIG8 , the memory 803, processor 802, and transceiver 801 are connected via bus 804. The bus is represented by a bold line in FIG8 . The connection between other components is for illustrative purposes only and is not intended to be limiting. The bus can be divided into an address bus, a data bus, a control bus, and so on. For ease of illustration, FIG8 shows only one bold line, but this does not mean that there is only one bus or only one type of bus.

[0338] In the embodiments of the present application, the processor may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc., and may implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of the present application may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor, etc.

[0339] In the embodiment of the present application, memory may include but is not limited to non-volatile memories such as hard disk drive (HDD) or solid-state drive (SSD), random access memory (RAM), erasable programmable read-only memory (EPROM), read-only memory (ROM) or portable read-only memory (CD-ROM), etc. Memory is any storage medium that can be used to carry or store program code in the form of instructions or data structures, and can be read and / or written by a computer (such as the computing resource management device shown in the present application), but is not limited thereto. The memory in the embodiment of the present application can also be a circuit or other arbitrarily capable of realizing a storage function, for storing program instructions and / or data.

[0340] The processor 802 is primarily used to process communication protocols and communication data, control the entire computing resource management device, execute software programs, and process software program data. The memory 803 is primarily used to store software programs and data. The transceiver 801 may include a control circuit and an antenna. The control circuit is primarily used to convert baseband signals into radio frequency signals and process radio frequency signals. The antenna is primarily used to transmit and receive radio frequency signals in the form of electromagnetic waves. Input and output devices, such as a touch screen, display, and keyboard, are primarily used to receive user input and output data to the user.

[0341] When the computing resource management device is turned on, the processor 802 can read the software program in the memory 803, interpret and execute the instructions of the software program, and process the data of the software program. When data needs to be sent wirelessly, the processor 802 performs baseband processing on the data to be sent and outputs the baseband signal to the radio frequency circuit. The radio frequency circuit performs radio frequency processing on the baseband signal and then transmits the radio frequency signal to the outside in the form of electromagnetic waves through the antenna. When data is sent to the computing resource management device, the radio frequency circuit receives the radio frequency signal through the antenna, converts the radio frequency signal into a baseband signal, and outputs the baseband signal to the processor 802. The processor 802 converts the baseband signal into data and processes the data.

[0342] In another implementation, the RF circuit and antenna may be provided independently of the processor performing baseband processing. For example, in a distributed scenario, the RF circuit and antenna may be remotely arranged independent of the computing resource management device.

[0343] The computing resource management device shown in the embodiment of the present application may also have more components than those shown in FIG8 , and the embodiment of the present application does not limit this. The method performed by the processor and transceiver shown above is only an example. For the specific steps performed by the processor and transceiver, please refer to the method described above.

[0344] In another possible implementation, in the computing resource management device shown in FIG7 , the processing module 701 may be one or more logic circuits, and the transceiver module 702 may be an input / output interface, also known as a communication interface, or an interface circuit, or an interface, etc. Alternatively, the transceiver module 702 may be a sending module and a receiving module, where the sending module may be an output interface and the receiving module may be an input interface, with the sending module and the receiving module integrated into a single module, such as an input / output interface. As shown in FIG9 , the computing resource management device 90 shown in FIG9 includes a logic circuit 901 and an interface 902. That is, the processing module 701 may be implemented using a logic circuit 901, and the transceiver module 702 may be implemented using an interface 902. The logic circuit 901 may be a chip, a processing circuit, an integrated circuit, or a system-on-chip (SoC) chip, etc., and the interface 902 may be a communication interface, an input / output interface, a pin, etc. For example, FIG9 uses the computing resource management device as a chip as an example, and the chip includes a logic circuit 901 and an interface 902.

[0345] In the embodiment of the present application, the logic circuit and the interface may also be coupled to each other. The embodiment of the present application does not limit the specific connection method of the logic circuit and the interface. For example, the logic circuit 901 can be used to perform the functions or steps implemented by the processing module 701 shown in Figure 7, and the interface 902 can be used to perform the functions or steps implemented by the transceiver module 702 shown in Figure 7. For a specific description of the logic circuit 901 and the interface 902, please refer to Figure 7 or the method embodiment shown above, and will not be described in detail here.

[0346] The above description of the computing resource management device is only an example. For the specific description of the computing resource management device shown in Figure 9, please refer to the above method embodiment or Figure 7 or Figure 8, which will not be described in detail here.

[0347] The computing resource management device shown in the embodiment of the present application can implement the method provided in the embodiment of the present application in the form of hardware, or can implement the method provided in the embodiment of the present application in the form of software, etc., and the embodiment of the present application does not limit this.

[0348] In the above embodiments, the description of the relevant steps and information can refer to the introduction of the above method embodiment, and will not be described in detail here. For the specific implementation of each embodiment shown in Figure 9, you can also refer to the above embodiments, and will not be described in detail here.

[0349] An embodiment of the present application further provides a computing resource management system, which includes a VNFM and an IM. The VNFM and IM interact to execute all or part of the steps in any of the aforementioned method embodiments.

[0350] In a possible implementation, the communication system further includes an NFVO, and the interaction between the VNFM, IM, and NFVO may be used to execute all or part of the steps in any of the aforementioned method embodiments.

[0351] In addition, the present application also provides a computer program, which is used to implement the operations and / or processing performed by each computing resource management device in the method provided by the present application.

[0352] The present application also provides a computer-readable storage medium, which stores computer code. When the computer code runs on a computer, the computer executes the operations and / or processing performed by each computing resource management device in the method provided by the present application.

[0353] The present application also provides a computer program product, which includes computer code or computer program. When the computer code or computer program is run on a computer, the operations and / or processes performed by the method provided in the present application are executed.

[0354] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or can be electrical, mechanical or other forms of connection.

[0355] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the technical effects of the solutions provided in the embodiments of the present application.

[0356] In addition, the functional modules in the various embodiments of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module. The above-mentioned integrated modules may be implemented in the form of hardware or software functional modules.

[0357] If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a readable storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned readable storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0358] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for managing computing resources, characterized in that: Applied to a virtualized network function manager (VNFM), the method includes: receiving an instantiation request for a virtualized network function (VNF) from a network function virtualization orchestrator (NFVO), wherein the instantiation request for the VNF includes an operating mode of the VNF; Determining addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the VNF and computing resource requirement information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources; A computing resource allocation request is sent to the infrastructure manager IM, wherein the computing resource allocation request includes the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF, and the computing resource allocation request is used to request allocation of resources for the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF.

2. The method according to claim 1, characterized in that The addressing information of the heterogeneous computing resource pool includes the identification information of the heterogeneous computing resource pool, and the computing resource requirement information of the VNF includes one or more of the VNF functional framework, the collaborative relationship of the computing units in the VNF functional framework, the communication bandwidth requirements between the computing units in the VNF functional framework, and the load balancing requirements between the computing units in the VNF functional framework.

3. The method according to claim 2, characterized in that The operation mode of the VNF includes a VNF inference mode or a VNF training mode.

4. The method according to claim 3, characterized in that The VNF functional framework includes a VNF reasoning framework or a VNF training framework.

5. The method according to any one of claims 1 to 4, characterized in that The heterogeneous computing resource pool is a computing resource pool stored in the network function virtualization infrastructure NFVI. The heterogeneous computing resource pool is a resource pool composed of at least two different types of computing resources among the central processing unit CPU, graphics processing unit GPU, digital signal processing DSP, application-specific integrated circuit ASIC and field programmable gate array FPGA.

6. The method according to any one of claims 1 to 5, characterized in that The instantiation request of the VNF also includes a VNF instance identifier, and the determining of addressing information of a heterogeneous computing resource pool corresponding to the operation mode of the VNF and computing resource requirement information of the VNF includes: Accessing the virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the instantiation request of the VNF; Based on the VNFD, addressing information of a heterogeneous computing resource pool corresponding to the operation mode of the VNF and computing resource requirement information of the VNF are determined.

7. The method according to any one of claims 1 to 5, characterized in that The determining of addressing information of a heterogeneous computing resource pool corresponding to the operation mode of the VNF includes: Sending a VNF lifecycle management permission request to the NFVO, wherein the VNF lifecycle management permission request is used to request to determine the addressing information of the heterogeneous computing resource pool corresponding to the operation mode of the VNF; Receive a VNF lifecycle management permission response from the NFVO, wherein the VNF lifecycle management permission response includes addressing information of the heterogeneous computing resource pool.

8. The method according to claim 7, characterized in that The instantiation request of the VNF also includes a VNF instance identifier, and determining computing resource requirement information of the VNF corresponding to the operation mode of the VNF includes: Accessing the virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the instantiation request of the VNF; Based on the VNFD, computing resource requirement information of the VNF corresponding to the operation mode of the VNF is determined.

9. The method according to any one of claims 1 to 8, characterized in that The method further comprises: receiving an update request for the VNF from the NFVO, wherein the update request for the VNF includes an updated operating mode of the VNF; Determining addressing information of an updated heterogeneous computing resource pool corresponding to the updated operating mode of the VNF and updated computing resource demand information of the VNF, wherein the updated heterogeneous computing resource pool includes at least two different types of computing resources; A computing resource allocation request is sent to the IM, wherein the computing resource allocation request includes the addressing information of the updated heterogeneous computing resource pool and the updated computing resource requirement information of the VNF, and the computing resource allocation request is used to request allocation of resources to the VNF in the updated heterogeneous computing resource pool according to the addressing information of the updated heterogeneous computing resource pool and the updated computing resource requirement information of the VNF.

10. A method for managing computing resources, characterized in that: Applied to an infrastructure manager IM, the method comprises: receiving a computing resource allocation request from a virtualized network function manager (VNFM), wherein the computing resource allocation request includes addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the virtualized network function (VNF) and computing resource demand information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources; Resources are allocated to the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource requirement information of the VNF.

11. The method according to claim 10, characterized in that After allocating resources for the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource requirement information of the VNF, the method further includes: Send a computing resource allocation response to the VNFM, wherein the computing resource allocation response is used to indicate that resource allocation is completed for the VNF in the heterogeneous computing resource pool.

12. The method according to claim 10 or 11, characterized in that The addressing information of the heterogeneous computing resource pool includes the identification information of the heterogeneous computing resource pool, and the computing resource requirement information of the VNF includes one or more of the VNF functional framework, the collaborative relationship of the computing units in the VNF functional framework, the communication bandwidth requirements between the computing units in the VNF functional framework, and the load balancing requirements between the computing units in the VNF functional framework.

13. The method according to claim 12, characterized in that The operation mode of the VNF includes a VNF inference mode or a VNF training mode.

14. The method according to claim 13, characterized in that The VNF functional framework includes a VNF reasoning framework or a VNF training framework.

15. The method according to any one of claims 10 to 14, characterized in that: The heterogeneous computing resource pool is a computing resource pool stored in the network function virtualization infrastructure NFVI. The heterogeneous computing resource pool is a resource pool composed of at least two different types of computing resources among the central processing unit CPU, graphics processing unit GPU, digital signal processing DSP, application-specific integrated circuit ASIC and field programmable gate array FPGA.

16. The method according to any one of claims 10 to 15, characterized in that: Before receiving the computing resource allocation request from the virtualized network function manager (VNFM), the method further includes: The heterogeneous computing resource pool is created in the network function virtualization infrastructure NFVI according to the computing resource pool description information, wherein the computing resource pool description information includes the computing resource type of the heterogeneous computing resource pool and the topological relationship of the nodes in the heterogeneous computing resource pool.

17. The method according to claim 16, characterized in that The topological relationship of the nodes in the heterogeneous computing resource pool is used to reflect the quantity and link connection relationship of at least two different types of computing resources in the heterogeneous computing resource pool.

18. The method according to claim 16 or 17, characterized in that After creating the heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the method further includes: Send a computing resource pool management notification, wherein the computing resource pool management notification includes the addressing information of the heterogeneous computing resource pool and the topological relationship of the nodes in the heterogeneous computing resource pool, and the computing resource pool management notification is used to notify the management of the heterogeneous computing resource pool.

19. The method according to claim 16 or 17, characterized in that Before creating the heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the method further includes: A computing resource pool creation request is received, wherein the computing resource pool creation request includes the computing resource pool description information.

20. The method according to claim 19, characterized in that After creating the heterogeneous computing resource pool in the network function virtualization infrastructure NFVI according to the computing resource pool description information, the method further includes: Send a computing resource pool creation response, wherein the computing resource pool creation response is used to indicate that the heterogeneous computing resource pool is successfully created.

21. The method according to any one of claims 10 to 20, characterized in that: The computing resource allocation request further includes addressing information of an updated heterogeneous computing resource pool corresponding to the updated operating mode of the VNF and updated computing resource demand information of the VNF, wherein the updated heterogeneous computing resource pool includes at least two different types of computing resources. The method further includes: Resources are allocated to the VNF in the updated heterogeneous computing resource pool according to the addressing information of the updated heterogeneous computing resource pool and the updated computing resource requirement information of the VNF.

22. A method for managing computing resources, characterized in that: The method comprises: The virtualized network function manager VNFM receives an instantiation request of a virtualized network function VNF, wherein the instantiation request of the VNF includes an operation mode of the VNF; The VNFM determines addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the VNF and computing resource demand information of the VNF, wherein the heterogeneous computing resource pool includes at least two different types of computing resources; The VNFM sends a computing resource allocation request to the infrastructure manager IM, wherein the computing resource allocation request includes addressing information of the heterogeneous computing resource pool and computing resource requirement information of the VNF; The IM receives the computing resource allocation request; The IM allocates resources to the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource demand information of the VNF.

23. The method according to claim 22, characterized in that The method further comprises: The network function virtualization orchestrator NFVO sends an instantiation request of the VNF to the VNFM.

24. The method according to claim 23, wherein The VNF instantiation request further includes a VNF instance identifier, and the VNFM determines addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the VNF and computing resource requirement information of the VNF, including: The VNFM accesses the virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the instantiation request of the VNF; The VNFM determines, based on the VNFD, addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the VNF and computing resource requirement information of the VNF.

25. The method according to claim 23, characterized in that The VNFM determines addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the VNF, including: The VNFM sends a VNF lifecycle management permission request to the NFVO, wherein the VNF lifecycle management permission request is used to request to determine addressing information of a heterogeneous computing resource pool corresponding to an operation mode of the VNF; The NFVO sends a VNF lifecycle management permission response to the VNFM, wherein the VNF lifecycle management permission response includes addressing information of the heterogeneous computing resource pool.

26. The method according to claim 25, characterized in that The VNF instantiation request further includes a VNF instance identifier, and the VNFM determines computing resource requirement information of the VNF corresponding to the operation mode of the VNF, including: The VNFM accesses the virtualized network function descriptor file VNFD of the VNF according to the VNF instance identifier in the instantiation request of the VNF; The VNFM determines computing resource requirement information of the VNF corresponding to the operation mode of the VNF based on the VNFD.

27. The method according to any one of claims 23 to 26, characterized in that After the IM receives the computing resource allocation request and allocates resources to the VNF in the heterogeneous computing resource pool according to the addressing information of the heterogeneous computing resource pool and the computing resource requirement information of the VNF, the method further includes: The IM sends a computing resource allocation response to the VNFM, wherein the computing resource allocation response is used to indicate that resource allocation is completed for the VNF in the heterogeneous computing resource pool; The VNFM sends an instantiation response of the VNF to the NFVO, wherein the instantiation response of the VNF is used to indicate that the VNF has completed the instantiation.

28. The method according to claim 27, characterized in that The method further comprises: The NFVO sends an update request of the VNF to the VNFM, wherein the update request of the VNF includes an updated operating mode of the VNF; The VNFM determines addressing information of an updated heterogeneous computing resource pool corresponding to the updated operating mode of the VNF and updated computing resource demand information of the VNF, wherein the updated heterogeneous computing resource pool includes at least two different types of computing resources; The VNFM sends a computing resource allocation request to the IM, wherein the computing resource allocation request includes updated addressing information of the heterogeneous computing resource pool and updated computing resource requirement information of the VNF; The IM allocates resources to the VNF in the updated heterogeneous computing resource pool based on the addressing information of the updated heterogeneous computing resource pool and the updated computing resource requirement information of the VNF.

29. A computing resource management device, characterized in that The device includes a module for executing the method according to any one of claims 1 to 9; or, the computing resource management device includes a processor, and the processor is used to execute the method according to any one of claims 1 to 9.

30. A computing resource management device, characterized in that: The device includes a module for executing the method according to any one of claims 10 to 21; or, the device includes a processor, and the processor is configured to execute the method according to any one of claims 10 to 21.

31. A computing resource management system, characterized in that: The system includes a virtualized network function manager VNFM and an infrastructure manager IM, the VNFM is the device according to claim 29, and the IM is the device according to claim 30.

32. A computing resource management device, characterized in that comprising a logic circuit and an interface, wherein the logic circuit and the interface are coupled; The interface is used to input and / or output information, and the logic circuit is used to execute the method according to any one of claims 1 to 21.

33. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store a computer program. When the computer program is executed, the method according to any one of claims 1 to 21 is executed.

Images