Resource management method, electronic equipment and storage medium

Abstract

The invention relates to the field of artificial intelligence, and provides a resource management method, electronic equipment and a storage medium, and the method comprises the steps: obtaining equipment information of computing equipment deployed at a node, and reporting the equipment information to a cluster end, so as to trigger the cluster end to create equipment resources based on the equipment information; registering virtual equipment information and reporting the virtual equipment information to a cluster end; when the resource management request is monitored, querying a resource deployment unit corresponding to the resource management request, determining target device information from device binding resources of the resource deployment unit, and mounting a computing device corresponding to the target device information to the resource deployment unit; the device binding resource is created by the cluster end based on the resource management request and the virtual device information, and the target device information is device information of a target device resource selected by the cluster end from unbound device resources based on the device binding resource. According to the method, the device and the storage medium provided by the invention, fine-grained resource management is realized, and the performance and efficiency of resource management are improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a resource management method, electronic equipment and storage medium. Background Art

[0002] In a large-scale cluster environment, thousands or even tens of thousands of computing devices are often deployed, and these thousands of computing devices may come from different manufacturers or belong to different product models.

[0003] To achieve dynamic allocation and management of computing device resources, in the K8S (Kubernetes) cluster, the device plugin can be used to report the vendor and quantity of computing devices, thereby providing a basic resource view for the cluster. Accordingly, the cluster can specify the vendor and quantity of computing devices to be used.

[0004] However, computing devices of different models from the same manufacturer may have different performances, which will directly affect the performance and efficiency of resource applications. How to achieve more fine-grained resource management has become an urgent problem to be solved. Summary of the invention

[0005] The present invention provides a resource management method, an electronic device and a storage medium, which are used to solve the defects in the related art that the resource management granularity is coarse and affects the resource application performance and efficiency.

[0006] The present invention provides a resource management method, comprising: Acquire device information of a computing device deployed at a node, and report the device information to a cluster end, so as to trigger the cluster end to create device resources of the computing device based on the device information; Register the virtual device information of the node and report it to the cluster end; In the case of monitoring a resource management request, querying a resource deployment unit corresponding to the resource management request, determining target device information from device binding resources of the resource deployment unit, and mounting a computing device corresponding to the target device information to the resource deployment unit; The device binding resource is created by the cluster end based on the resource management request and the virtual device information, and the target device information is device information corresponding to the target device resource selected by the cluster end from all unbound device resources based on the device binding resource; The registering the virtual device information of the node and reporting it to the cluster end includes: The node agent component of the node is based on registering virtual device information to trigger the node agent component to report the virtual device information to the cluster end, and the number of virtual devices included in the virtual device information is greater than or equal to the number of computing devices deployed at the node.

[0007] According to a resource management method provided by the present invention, the step of obtaining device information of a computing device deployed at a node includes: Scan the PCI device at the node to obtain the device model of the PCI device; Based on the device model of the PCI device, device information of a computing device corresponding to the PCI device is determined.

[0008] According to a resource management method provided by the present invention, the determining of device information of a computing device corresponding to the PCI device based on the device model of the PCI device includes: Determine, based on the device model of the PCI device, the device model of the computing device corresponding to the PCI device; Based on the correspondence between the general device model and the subdivided device model, the general device model corresponding to the device model of the computing device is determined, and the device information is determined based on the general device model.

[0009] According to a resource management method provided by the present invention, the device resources include device information, performance information and status information of the computing device.

[0010] According to a resource management method provided by the present invention, the querying of the resource deployment unit corresponding to the resource management request includes: Receiving a virtual device allocation request sent by a node agent component of the node; Based on the virtual device identifier corresponding to the virtual device allocation request, the resource deployment unit corresponding to the resource management request is queried.

[0011] The present invention also provides a resource management method, comprising: Receiving device information of a computing device deployed at a node and virtual device information of the node, and creating device resources of the computing device based on the device information, wherein the device information is obtained by a device agent component at the node, the virtual device information is registered and reported based on the node agent component of the node, and the number of virtual devices included in the virtual device information is greater than or equal to the number of computing devices deployed at the node; In case of receiving a resource management request, creating a device binding resource of a resource deployment unit corresponding to the resource management request based on the resource management request and the virtual device information; Based on the device binding resources, a target device resource is selected from all unbound device resources, and the target device information in the device binding resources is updated based on the device information corresponding to the target device resource, so that when the device proxy component listens to the resource management request, it determines the target device information from the device binding resources of the resource deployment unit corresponding to the resource management request, and mounts the computing device corresponding to the target device information to the resource deployment unit.

[0012] According to a resource management method provided by the present invention, the device resources include device information, performance information and status information of the computing device.

[0013] According to a resource management method provided by the present invention, the step of selecting a target device resource from all unbound device resources based on the device binding resource includes: Based on the status information of all device resources, determine the unbound device resources; Based on the device-bound resources, and the device information and performance information of the unbound device resources, a target device resource is selected from all unbound device resources.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the resource management methods described above when executing the program.

[0015] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the resource management method described in any one of the above is implemented.

[0016] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the resource management methods described above.

[0017] The resource management method, electronic device and storage medium provided by the present invention obtain device information of computing devices deployed at nodes and report it to the cluster side to create device resources. In addition, device binding resources are created on the cluster side based on resource management requests, thereby combining device resources and device binding resources to achieve fine-grained resource management and allocation, thereby improving the performance and efficiency of resource management in heterogeneous clusters. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 It is a flowchart of resource management based on K8S cluster in related technologies; Figure 2 It is one of the flow charts of the resource management method provided by the present invention; Figure 3 This is the second flow chart of the resource management method provided by the present invention; Figure 4 This is the third flow chart of the resource management method provided by the present invention; Figure 5 It is one of the structural schematic diagrams of the resource management device provided by the present invention; Figure 6 This is the second structural diagram of the resource management device provided by the present invention; Figure 7 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In a large-scale cluster environment, thousands or even tens of thousands of computing devices are often deployed. These thousands of computing devices may come from different manufacturers, or belong to different product models of the same manufacturer.

[0022] Here, the computing device may specifically be an artificial intelligence chip, such as a GPU (Graphics Processing Unit), or may be any one or more of a TPU (Tensor Processing Unit), an NPU (Neural network Processing Unit), a DPU (Deep learning Processing Unit), an APU (Accelerated Processing Unit), and a GPGPU (General-Purpose computing on Graphics Processing Unit), and the embodiments of the present invention do not specifically limit this.

[0023] K8S cluster is an open source cloud server management platform. Taking the computing device as GPU as an example, in order to realize the dynamic allocation and management of GPU resources, in the K8S cluster, the manufacturer and quantity of GPU can be reported through the device plug-in, thereby providing a basic resource view for the cluster.

[0024] For example, Figure 1 This is a flowchart of resource management based on K8S cluster in related technologies. Figure 1 As shown in the figure, GPUs from two manufacturers, manufacturer A and manufacturer B, are deployed at node 1 and node 2, respectively. The GPUs from manufacturer A include GPU1-GPU4, and the GPUs from manufacturer B include GPU5-GPU8. For the GPUs from manufacturer A, manufacturer A provides a device plug-in for the K8S cluster; in addition, for the GPUs from manufacturer B, manufacturer B also provides a device plug-in for the K8S cluster. Therefore, the device plug-in from manufacturer A is installed at node 1, and the device plug-in from manufacturer B is installed at node 2, so that the K8S cluster can identify the GPUs from manufacturer A and manufacturer B.

[0025] The resource management process can be divided into the following five steps: Step ①: Each manufacturer's device plug-in reports the number of GPUs belonging to its own manufacturer to the node agent component (also known as Kubelet) at the node to which it belongs. For example, at node 1, the information reported by the device plug-in of manufacturer A can be A.com / GPU:4.

[0026] Step ②: The node agent component of each node reports the number of GPUs of each manufacturer to the API (Application Programming Interface) server (also known as API Server).

[0027] Step 3: The API server receives the Pod creation request sent by the user. The Pod creation request here can carry the required resources. For example, if two GPUs from manufacturer B are required, they can be identified as B.com / GPU:2. The Pod here is the smallest deployment unit in the K8S cluster, which can represent a container or a combination of multiple containers.

[0028] Step 4: The K8S scheduler (also known as K8S Scheduler) monitors the creation of the Pod and schedules the Pod to Node 2 based on the resources required for the Pod creation.

[0029] Step 5: The node agent component of node 2 interacts with the device plug-in of vendor B required for Pod creation, and allocates two GPUs from vendor B's GPUs to the Pod.

[0030] It can be seen that in the above solution, if there are GPUs from multiple manufacturers in the K8S cluster, it is necessary to install the device plug-in of each manufacturer to facilitate the corresponding GPU identification.

[0031] In addition, since GPUs of different product models from the same manufacturer may also have differences in performance, simply describing the manufacturer's GPU quantity without reporting the specific GPU model cannot truly reflect the performance of each GPU. For example, manufacturer A has two GPUs of different models, and the parameters and performance of the two GPUs are completely different, but manufacturer A's device plug-in can only report the two GPUs deployed by manufacturer A, in a format similar to A.com / GPU:2. Accordingly, since the cluster cannot obtain the specific GPU model, when performing resource management, the cluster can only specify the manufacturer and number of GPUs to be used, but cannot specify the specific GPU model. This results in the current resource management method being coarse-grained, which directly affects the performance and efficiency of resource applications.

[0032] In view of the above problems, an embodiment of the present invention provides a resource management method. Figure 2 It is one of the flow charts of the resource management method provided by the present invention, such as Figure 2 As shown, the method is applied to the nodes of a heterogeneous cluster, specifically to the device agent (also known as Device Agent) component at the node. The heterogeneous cluster here refers to a computing cluster composed of computing devices from different manufacturers or different product models of the same manufacturer. The method includes: Step 210: Obtain device information of a computing device deployed at a node, and report the device information to a cluster end to trigger the cluster end to create device resources of the computing device based on the device information.

[0033] Specifically, in a heterogeneous cluster, a node is a physical server in the heterogeneous cluster, which carries the computing resources required for the operation of the heterogeneous cluster. One or more computing devices can be deployed on each node.

[0034] Furthermore, in the embodiment of the present invention, for any node, each computing device deployed at the node may come from different manufacturers, or belong to different product models of the same manufacturer. Alternatively, for any node, each computing device deployed at the node may come from the same manufacturer, or belong to the same product model of the same manufacturer, and for different nodes, the computing devices deployed at different nodes may come from different manufacturers, or belong to different product models of the same manufacturer.

[0035] The device proxy component can be deployed at the node of a heterogeneous cluster, and the device proxy component can detect all computing devices deployed at the node, thereby obtaining the device information of all computing devices deployed at the node. It should be noted that the device proxy component is a component set at the node, which is different from the node proxy component (also known as Kubelet). The device proxy component can communicate with the node proxy component. In an embodiment of the present invention, the device proxy component is a newly added component at each node in a heterogeneous cluster, which is used to obtain the device information of the computing device deployed at the node, and cooperate with the node proxy component to exchange information with the cluster end.

[0036] The device information of the computing device deployed at the node can be obtained by scanning the PCI (Peripheral Component Interconnect) device at the node; or, it can be achieved by integrating the communication interfaces of computing devices from various manufacturers, which is not specifically limited in the embodiments of the present invention.

[0037] The device information thus obtained may include information identifying the computing device, specifically the device ID (Identity document) of the computing device. In addition, the device information may also include the manufacturer information of the computing device, such as the device vendor ID of the computing device. Alternatively, the device information may also include information such as the node name and PCI address of the computing device.

[0038] After obtaining the device information of each computing device deployed at the node, the device proxy component can report the device information of each computing device to the cluster end. The cluster end referred to here, that is, the resource management end of the heterogeneous cluster, can be, for example, a K8S cluster, which can specifically include the API server, scheduler, etc. in the K8S cluster, and the embodiment of the present invention does not specifically limit this.

[0039] Accordingly, after receiving the device information of the computing device at the node reported by the device proxy component at the node, the cluster end can create device resources of the computing device deployed at the node based on the device information.

[0040] The device resource here is a custom resource. For example, if the computing device is a GPU, the device resource can be a GPU CRD (Custom Resource Definitions). It can be understood that the device resources maintained on the cluster side can persistently and finely reflect the situation of each computing device deployed at the node, which provides a richer resource view for the cluster side compared to the related technology that only reports the number of GPUs of each manufacturer.

[0041] Step 220: register the virtual device information of the node and report it to the cluster end.

[0042] Specifically, in order to continue to use the original resource management logic of the K8S cluster and avoid the large-scale workload caused by modifying the resource management logic of the K8S cluster, the device proxy component can also register virtual device information and report it to the cluster end. The virtual device information here refers to the information obtained by the virtual computing device through device registration for the node. The virtual device information can indicate that a large number of computing devices from any manufacturer are deployed at the node. For example, the virtual device information can be recorded as FakeDevice, and the specific form is example.com / FakeDevice:1000, where 1000 is the number of virtual computing devices in the example, and the actual value can be any number greater than or equal to the number of computing devices actually deployed at the node. By setting a number that is much larger than the number of computing devices actually deployed, overselling can be achieved to improve the resource utilization of the K8S cluster.

[0043] Accordingly, the cluster end can receive the virtual device information of the node.

[0044] Step 230, when a resource management request is monitored, query the resource deployment unit corresponding to the resource management request, determine the target device information from the device binding resources of the resource deployment unit, and mount the computing device corresponding to the target device information to the resource deployment unit; The device binding resource is created by the cluster end based on the resource management request and the virtual device information, and the target device information is device information corresponding to the target device resource selected by the cluster end from all unbound device resources based on the device binding resource.

[0045] Specifically, in the resource management scenario, the user end can send a resource management request to the cluster end. Here, the user end is used to realize the interaction between the user and the heterogeneous cluster. The user end can interact with the cluster end. For example, the user end can interact with the API server in the cluster end to realize functions such as querying the cluster status and managing cluster resources, and then realize the use of computing resources provided by the computing devices at each node through the cluster end.

[0046] The resource management request here can be a Pod creation request, that is, a request for creating a Pod and configuring computing resources for the created Pod. The Pod here is the smallest deployment unit in the K8S cluster, which can represent a container or a combination of multiple containers. In an embodiment of the present invention, the Pod created based on the resource management request is recorded as a resource deployment unit.

[0047] Resource management requests often carry the user's requirements for the computing resources that the created Pod expects to configure, such as the number and manufacturer of computing devices to be configured. Therefore, after receiving the resource management request, the cluster can create a corresponding resource deployment unit based on the resource management request, and configure the corresponding computing resources for the resource deployment unit based on the requirements for computing resources in the resource management request and the virtual device information reported by the node. Here, the computing resources configured for the resource deployment unit are recorded as device-bound resources.

[0048] Here, the device binding resource is a custom resource. For example, when the computing device is a GPU, the device binding resource can be specifically a GPUBinding CRD. It is understandable that the device binding resource maintained on the cluster side can persistently describe the binding relationship between the resource deployment unit and the computing device.

[0049] Since the virtual device information is the virtual device information registered by the device proxy component, configuring the device binding resource for the resource deployment unit can only realize the binding of the resource deployment unit with the node, but does not realize the real binding with the computing device deployed in the node. Therefore, the cluster side also needs to select the target device resources that meet the requirements of the computing resources expected to be configured by the resource management request from all the device resources deployed at the node that have not yet been bound, that is, from all the device resources that have not yet been bound, select the target device resources corresponding to the requirements put forward by the device binding resource, and write the device information corresponding to the target device resource into the device binding resource. At this time, the device binding resource contains the device information of the real computing device that is actually bound to the resource deployment unit, that is, the target device information.

[0050] Thus, the device binding resource may include the name of the resource deployment unit that needs to use the computing device, the actual container name, and the name of the device resource that needs to be bound. In addition, when the real binding with the computing device has not yet been achieved, the name of the device resource that needs to be bound contained in the device binding resource is empty; and when the computing device that needs to be bound is determined, the name of the device resource that needs to be bound contained in the device binding resource can be expressed as a node identifier and the serial number of the computing device deployed on the node, such as node1-0, which means binding the 0th card on the node node1.

[0051] In addition, the device proxy component at the node can also monitor the above-mentioned resource management request. And, in the case of determining that the resource management request is monitored, the device proxy component can query the resource deployment unit corresponding to the resource management request, thereby querying the device binding resource of the resource deployment unit and reading the target device information from the device binding resource. That is, the device proxy component can determine the computing device deployed at the node that needs to be bound to the resource deployment unit through the target device information in the device binding resource, that is, the computing device corresponding to the target device information.

[0052] Therefore, after determining the computing device corresponding to the target device information, the computing device can be mounted to the resource deployment unit to support the operation of the resource deployment unit.

[0053] In the method provided in an embodiment of the present invention, device information of a computing device deployed at a node is obtained and reported to a cluster end to create device resources. In addition, device binding resources are created on the cluster end based on a resource management request, thereby combining device resources and device binding resources to achieve fine-grained resource management and allocation, thereby improving the performance and efficiency of resource management in a heterogeneous cluster.

[0054] Based on the above embodiment, in step 210, obtaining device information of a computing device deployed at a node includes: Scan the PCI device at the node to obtain the device model of the PCI device; Based on the device model of the PCI device, device information of a computing device corresponding to the PCI device is determined.

[0055] Specifically, compared to the related art, which requires configuring device plug-ins of corresponding manufacturers for GPUs of different manufacturers, in the embodiments of the present invention, it is only necessary to install a device proxy component at the node to replace the device plug-ins of each manufacturer. That is, no matter how many computing devices of different manufacturers are deployed at the node, only one device proxy component needs to be installed. That is, in the embodiments of the present invention, the device proxy component can obtain the device information of all computing devices deployed at the node.

[0056] In order to obtain device information of computing devices from different manufacturers, the device proxy component can scan the PCI devices at the node. It can be understood that computing devices from various manufacturers need to be connected to the node through the PCI bus, and the PCI device here is the computing device connected through the PCI bus. That is, the device proxy component can obtain all computing devices deployed at the node by scanning the PCI devices at the node.

[0057] By scanning the PCI device, the device model of the PCI device can be obtained, for example, the VendorID (vendor identifier) ​​and Device ID (device identification) of the PCI device. It can be understood that in the device model of the PCI device, the Vendor ID can reflect the manufacturer of the computing device, and the Device ID can reflect the model of the computing device. For example, the device model is [10de:2503], where 10de is the Vendor ID, representing a certain manufacturer, and 2503 is the device number under the manufacturer; 1ee0 in [1ee0: 0003] represents another manufacturer, and 0003 represents the device number under the manufacturer. Therefore, after obtaining the device model of the PCI device, the manufacturer, model and other device information of the computing device corresponding to the PCI device can be determined based on this.

[0058] In the method provided in the embodiment of the present invention, by scanning the PCI devices at the node, unified acquisition of device information of computing devices of various manufacturers deployed at the node is achieved, without the need to install the respective device plug-ins of each manufacturer separately, which greatly improves the convenience of resource management.

[0059] Furthermore, since the device information carries the device model, resource requirements for using a specific model can be met in resource management.

[0060] Based on any of the above embodiments, in step 210, determining the device information of the computing device corresponding to the PCI device based on the device model of the PCI device includes: Determine, based on the device model of the PCI device, the device model of the computing device corresponding to the PCI device; Based on the correspondence between the general device model and the subdivided device model, the general device model corresponding to the device model of the computing device is determined, and the device information is determined based on the general device model.

[0061] Specifically, in the device model of the PCI device, the Device ID corresponds to the device model of the computing device. For example, when the device model of the PCI device is [10de:2503], 2503 is the Device ID, and the device model of the corresponding computing device can be determined based on 2503; when the device model of the PCI device is [1ee0: 0003], 0003 is the Device ID, and the device model of the corresponding computing device can be determined based on 0003. Thus, after obtaining the device model of the PCI device, the device model of the corresponding computing device can be obtained.

[0062] In actual applications, a manufacturer may have multiple models of computing devices under the same series. These computing devices are of different models and have slightly different performance parameters, but when users use them, they are all regarded as the same type of computing devices for resource allocation and management.

[0063] In order to facilitate the unified management of similar computing devices, the correspondence between general device models and detailed device models can be pre-built. Among them, the general device model is the device model commonly used to display to users or that users are accustomed to using, such as the serial number of a type of computing device or an abbreviated model; the detailed device model is the detailed, unabbreviated device model of the computing device.

[0064] It can be understood that the device model of the computing device obtained based on the device model of the PCI device is usually a subdivided device model. Therefore, based on the correspondence between the general device model and the subdivided device model, the device model of the computing device can be mapped to the corresponding general device model, and the device information of the computing device can be determined based on this general device model, thereby realizing unified management of a class of computing devices.

[0065] Based on any of the above embodiments, the device resources include device information, performance information and status information of the computing device.

[0066] Specifically, in addition to describing the device information of the computing device, the device resources may also describe the performance information and status information of the computing device.

[0067] The device information of the computing device may include the device ID, and may also include the manufacturer ID, node name, PCI address, etc. The performance information of the computing device may include the size of the video memory, power, CPU affinity, etc. The status information of the computing device may indicate whether the computing device is allocated for use, and may also include the health status of the computing device, etc. For example, when the computing device is allocated for use, the status information of the computing device may include information of the allocated resource deployment unit, such as the name and namespace of the allocated resource deployment unit, and the ID of the user corresponding to the resource deployment unit, etc.

[0068] Accordingly, when the cluster side selects target device resources from all unbound device resources based on device binding resources, it can specifically determine the unbound device resources from all device resources based on the status information of all device resources, and then select the target device resources that match the requirements of the device binding resources based on the device information and performance information of all unbound device resources.

[0069] Based on any of the above embodiments, in step 220, registering the virtual device information of the node and reporting it to the cluster end includes: The node agent component of the node is based on registering virtual device information to trigger the node agent component to report the virtual device information to the cluster end, and the number of virtual devices included in the virtual device information is greater than or equal to the number of computing devices deployed at the node.

[0070] Specifically, a node agent component (also known as Kubelet) is deployed at the node. The device agent component can communicate with the node agent component. Specifically, the node's virtual device information can be registered through the node agent component, and then the node agent component reports the virtual device information to the cluster end, thereby realizing the registration and reporting of virtual device information.

[0071] Here, when the number of virtual devices included in the virtual device information is greater than the number of computing devices actually deployed at the node, overselling can be achieved to improve the resource utilization of the K8S cluster.

[0072] Based on any of the above embodiments, in step 230, querying the resource deployment unit corresponding to the resource management request includes: Receiving a virtual device allocation request sent by a node agent component of the node; Based on the virtual device identifier corresponding to the virtual device allocation request, the resource deployment unit corresponding to the resource management request is queried.

[0073] Specifically, at the node, the node agent component can monitor the resource management request, and when the node agent component monitors the resource management request, the node agent component can send a virtual device allocation request to the device agent component. Here, the virtual device allocation request is used to request the device agent component to allocate a virtual device to meet the requirements of the resource management request.

[0074] Accordingly, the device proxy component can receive the virtual device allocation request sent by the node proxy component, and after receiving the virtual device allocation request, the device identifier of the virtual device to be allocated, i.e., the virtual device identifier, can be obtained therefrom. Since the device binding resource is created on the cluster side and based on the resource management request and the virtual device information, a binding relationship between the resource management request and the allocated virtual device identifier is established on the cluster side. Based on this, the device proxy component can query the resource deployment unit corresponding to the resource management request bound to the virtual device identifier according to the virtual device identifier in the virtual device allocation request, thereby realizing the query of the resource deployment unit, and further providing conditions for realizing the computing device mounting for the resource deployment unit.

[0075] Based on any of the above embodiments, Figure 3 This is the second flow chart of the resource management method provided by the present invention, such as Figure 3 As shown, the method is applied to the cluster end of a heterogeneous cluster. The cluster end referred to here, that is, the resource management end of the heterogeneous cluster, can be, for example, a K8S cluster, and can specifically include an API server, a scheduler, etc. in the K8S cluster. The embodiment of the present invention does not make specific limitations on this.

[0076] The method includes: Step 310, receiving device information of a computing device deployed at a node and virtual device information of the node, and creating device resources of the computing device based on the device information, wherein the device information is obtained by a device proxy component at the node.

[0077] Specifically, a device agent (also known as Device Agent) component can be deployed at the node of a heterogeneous cluster. The device agent component can detect all computing devices deployed at the node, thereby obtaining the device information of all computing devices deployed at the node. It should be noted that the device agent component is a component set at the node, which is different from the node agent component (also known as Kubelet). The device agent component can communicate with the node agent component. In an embodiment of the present invention, the device agent component is a newly added component at each node in a heterogeneous cluster, which is used to obtain the device information of the computing device deployed at the node, and cooperate with the node agent component to exchange information with the cluster end.

[0078] The device proxy component can obtain device information of computing devices deployed at the node, which can be achieved by scanning PCI devices at the node; or, it can be achieved by integrating communication interfaces of computing devices from various manufacturers, which is not specifically limited in the embodiments of the present invention.

[0079] The device information thus obtained may include information identifying the computing device, specifically, the device ID of the computing device. In addition, the device information may also include the manufacturer information of the computing device, such as the device vendor ID of the computing device. Alternatively, the device information may also include information such as the node name and PCI address of the computing device.

[0080] After obtaining the device information of each computing device deployed at the node, the device proxy component can report the device information of each computing device to the cluster end.

[0081] Accordingly, after receiving the device information of the computing device at the node reported by the device proxy component at the node, the cluster end can create device resources of the computing device deployed at the node based on the device information.

[0082] The device resource here is a custom resource. For example, if the computing device is a GPU, the device resource can be a GPU CRD. It can be understood that the device resources maintained on the cluster side can persistently and finely reflect the situation of each computing device deployed at the node, which provides a richer resource view for the cluster side compared to the related technology that only reports the number of GPUs of each manufacturer.

[0083] In addition, in order to continue to use the original resource management logic of the K8S cluster and avoid the large-scale workload caused by modifying the resource management logic of the K8S cluster, the device proxy component can also register virtual device information and report it to the cluster end. The virtual device information here refers to the information obtained by the virtual computing device through device registration for the node. The virtual device information can indicate that a large number of computing devices from any manufacturer are deployed at the node. For example, the virtual device information can be recorded as FakeDevice, and the specific form is example.com / FakeDevice:1000, where 1000 is the number of virtual computing devices in the example, and the actual value can be any number greater than or equal to the number of computing devices actually deployed at the node. By setting a number that is much larger than the number of computing devices actually deployed, overselling can be achieved to improve the resource utilization of the K8S cluster.

[0084] Accordingly, the cluster end can receive the virtual device information of the node.

[0085] Step 320: When a resource management request is received, a device binding resource of a resource deployment unit corresponding to the resource management request is created based on the resource management request and the virtual device information.

[0086] Specifically, in the resource management scenario, the user end can send a resource management request to the cluster end. Here, the user end is used to realize the interaction between the user and the heterogeneous cluster. The user end can interact with the cluster end. For example, the user end can interact with the API server in the cluster end to realize functions such as querying the cluster status and managing cluster resources, and then realize the use of computing resources provided by the computing devices at each node through the cluster end.

[0087] The resource management request here can be a Pod creation request, that is, a request for creating a Pod and configuring computing resources for the created Pod. The Pod here is the smallest deployment unit in the K8S cluster, which can represent a container or a combination of multiple containers. In an embodiment of the present invention, the Pod created based on the resource management request is recorded as a resource deployment unit.

[0088] Resource management requests often carry the user's requirements for the computing resources that the created Pod expects to configure, such as the number and manufacturer of computing devices to be configured. Therefore, after receiving the resource management request, the cluster can create a corresponding resource deployment unit based on the resource management request, and configure the corresponding computing resources for the resource deployment unit based on the requirements for computing resources in the resource management request and the virtual device information reported by the node. Here, the computing resources configured for the resource deployment unit are recorded as device-bound resources.

[0089] Here, the device binding resource is a custom resource. For example, when the computing device is a GPU, the device binding resource can be specifically a GPUBinding CRD. It is understandable that the device binding resource maintained on the cluster side can persistently describe the binding relationship between the resource deployment unit and the computing device.

[0090] Step 330, based on the device binding resources, select the target device resources from all unbound device resources, and update the target device information in the device binding resources based on the device information corresponding to the target device resources, so that when the device proxy component listens to the resource management request, it can determine the target device information from the device binding resources of the resource deployment unit corresponding to the resource management request, and mount the computing device corresponding to the target device information to the resource deployment unit.

[0091] Specifically, since the virtual device information is virtual device information registered by the device proxy component, configuring the device binding resource for the resource deployment unit can only realize the binding of the resource deployment unit with the node, but does not realize the real binding with the computing device deployed in the node. Therefore, the cluster side also needs to select the target device resources that meet the requirements of the computing resources expected to be configured by the resource management request from all the device resources deployed at the node that have not yet been bound, that is, from all the device resources that have not yet been bound, select the target device resources corresponding to the requirements put forward by the device binding resource, and write the device information corresponding to the target device resource into the device binding resource. At this time, the device information of the real computing device actually bound to the resource deployment unit, that is, the target device information, is written into the device binding resource.

[0092] The device proxy component at the node can also monitor the resource management request mentioned above. In addition, when it is determined that a resource management request is monitored, the device proxy component can query the resource deployment unit corresponding to the resource management request, thereby querying the device binding resource of the resource deployment unit and reading the target device information from the device binding resource. That is, the device proxy component can determine the computing device deployed at the node that needs to be bound to the resource deployment unit through the target device information in the device binding resource, that is, the computing device corresponding to the target device information.

[0093] Therefore, after determining the computing device corresponding to the target device information, the computing device can be mounted to the resource deployment unit to support the operation of the resource deployment unit.

[0094] In the method provided in an embodiment of the present invention, device information of a computing device deployed at a node is obtained and reported to a cluster end to create device resources. In addition, device binding resources are created on the cluster end based on a resource management request, thereby combining device resources and device binding resources to achieve fine-grained resource management and allocation, thereby improving the performance and efficiency of resource management.

[0095] Based on any of the above embodiments, the device resources include device information, performance information and status information of the computing device.

[0096] Specifically, in addition to describing the device information of the computing device, the device resources may also describe the performance information and status information of the computing device.

[0097] The device information of the computing device may include the device ID, and may also include the manufacturer ID, node name, PCI address, etc. The performance information of the computing device may include the size of the video memory, power, CPU affinity, etc. The status information of the computing device may indicate whether the computing device is allocated for use, and may also include the health status of the computing device, etc. For example, when the computing device is allocated for use, the status information of the computing device may include information of the allocated resource deployment unit, such as the name and namespace of the allocated resource deployment unit, and the ID of the user corresponding to the resource deployment unit, etc.

[0098] Accordingly, in step 330, the selecting of a target device resource from all unbound device resources based on the device binding resource includes: Based on the status information of all device resources, determine the unbound device resources; Based on the device-bound resources, and the device information and performance information of the unbound device resources, a target device resource is selected from all unbound device resources.

[0099] Specifically, when the cluster side selects the target device resources from all unbound device resources based on the device binding resources, it can determine the unbound device resources from all device resources based on the status information of all device resources, and then select the target device resources that match the requirements of the device binding resources based on the device information and performance information of all unbound device resources.

[0100] The target device resource selected here is a device resource that is not bound and whose device information and performance information match the requirements of the device-bound resource. For example, it can be a device resource whose model is consistent with the requirements of the device-bound resource, or a device resource whose video memory size meets the requirements of the device-bound resource. The embodiment of the present invention does not specifically limit this.

[0101] Based on any of the above embodiments, Figure 4 This is the third flow chart of the resource management method provided by the present invention, such as Figure 4 As shown, GPU1-GPU4 are deployed at node 1, and GPU5-GPU8 are deployed at node 2. GPU1-GPU4 can belong to the same manufacturer or different manufacturers, and GPU5-GPU8 can belong to the same manufacturer or different manufacturers. To facilitate the description of the steps in the resource management process, node 1 is taken as an example below: Step ①: The device proxy component at node 1 can detect the GPU device deployed on node 1, and cooperate with the configuration in the pre-set ConfigMap to obtain the device information of the detected GPU device, and report the device information to the API server on the cluster side.

[0102] Here, the GPU device deployed on node 1 is detected, which can be achieved by scanning the PCI device on node 1; In addition, ConfigMap can include the mapping relationship between the general device models and the sub-device models of computing devices of various manufacturers. Based on ConfigMap, the model of the GPU detected by the device proxy component can be mapped to the general device model, so that the device proxy component can generate device information based on the general device model. Furthermore, ConfigMap can include the manufacturer's general device model display_name and the sub-device model Device ID.

[0103] Step ②: The device agent component at node 1 can create a corresponding device resource GPU CRD for each device information through the API server on the cluster side. The device resource here can represent the specific information of the GPU deployed on node 1.

[0104] Step ③: The device agent component at node 1 can register the virtual device information through the node agent component at node 1, thereby utilizing the original K8S control mechanism to achieve resource management.

[0105] Step ④: The node agent component at node 1 can report the virtual device information to the API server at the cluster end.

[0106] Step 5. The API server on the cluster side receives the Pod creation request sent by the user side. The Pod creation request here can declare the required device model and quantity, for example, two computing devices of a certain model are required.

[0107] Step ⑥: The GPU Controller on the cluster side can declare the number of devices required based on the Pod creation request. Create the corresponding number of device binding resources GPUBinding CRD, and indicate the number of virtual devices required for the Pod corresponding to the Pod creation request. For example, if 2 GPUs are required, you can specify example.com / fakeDevice:2.

[0108] Step 7, the cluster-side scheduler GPU scheduler can search for suitable device resource GPU CRDs from all unbound device resource GPU CRDs according to the device binding resource GPUBindingCRD of the Pod and the scheduling strategy, and return the corresponding node name as the scheduling result. For example, the scheduling result can be node 1. Since node 1 declares in the virtual device information that it has a large number of virtual devices deployed, for example, 1,000 virtual devices deployed, it can be guaranteed that node 1 must meet the virtual device quantity conditions required by the Pod corresponding to the Pod creation request. The scheduling strategy here can be a pre-set strategy for matching demand and device resources.

[0109] Step ⑧, the node agent component on node 1 can monitor the creation of Pod, and after monitoring the creation of Pod, request the device agent component to allocate the number of virtual devices required by Pod. Accordingly, the device agent component can query the Pod of the requested device according to the ID of the virtual device requested for allocation, and then the device agent component can obtain the device binding resource GPUBinding CRD of the Pod through the API server on the cluster side, and find the GPU device that actually needs to be mounted according to the device binding resource GPUBindingCRD of the Pod, and return the device information of the GPU device to the node agent component, thereby mounting the GPU device to the Pod.

[0110] The resource management device provided by the present invention is described below. The resource management device described below and the resource management method described above can be referenced to each other.

[0111] Figure 5 is one of the structural diagrams of the resource management device provided by the present invention, such as Figure 5 As shown, the device can be applied to a device agent component at a node, and the device includes: A device information reporting unit 510, configured to obtain device information of a computing device deployed at a node, and report the device information to a cluster end, so as to trigger the cluster end to create device resources of the computing device based on the device information; A virtual information reporting unit 520, configured to register the virtual device information of the node and report it to the cluster end; The binding unit 530 is used to query the resource deployment unit corresponding to the resource management request when monitoring the resource management request, determine the target device information from the device binding resources of the resource deployment unit, and mount the computing device corresponding to the target device information to the resource deployment unit; The device binding resource is created by the cluster end based on the resource management request and the virtual device information, and the target device information is device information corresponding to the target device resource selected by the cluster end from all unbound device resources based on the device binding resource.

[0112] In the apparatus provided in an embodiment of the present invention, device information of a computing device deployed at a node is obtained and reported to a cluster end to create device resources. In addition, device binding resources are created on the cluster end based on a resource management request, thereby combining device resources and device binding resources to achieve fine-grained resource management and allocation, thereby improving the performance and efficiency of resource management in a heterogeneous cluster.

[0113] Based on any of the above embodiments, the device information reporting unit is specifically used for: Scan the PCI device at the node to obtain the device model of the PCI device; Based on the device model of the PCI device, device information of a computing device corresponding to the PCI device is determined.

[0114] Based on any of the above embodiments, the device information reporting unit is specifically used for: Determine, based on the device model of the PCI device, the device model of the computing device corresponding to the PCI device; Based on the correspondence between the general device model and the subdivided device model, the general device model corresponding to the device model of the computing device is determined, and the device information is determined based on the general device model.

[0115] Based on any of the above embodiments, the device resources include device information, performance information and status information of the computing device.

[0116] Based on any of the above embodiments, the virtual information reporting unit is specifically used for: The node agent component of the node is based on registering virtual device information to trigger the node agent component to report the virtual device information to the cluster end, and the number of virtual devices included in the virtual device information is greater than or equal to the number of computing devices deployed at the node.

[0117] Based on any of the above embodiments, the binding unit is specifically used for: Receiving a virtual device allocation request sent by a node agent component of the node; Based on the virtual device identifier corresponding to the virtual device allocation request, the resource deployment unit corresponding to the resource management request is queried.

[0118] Figure 6 This is the second structural diagram of the resource management device provided by the present invention, such as Figure 6As shown, the device can be applied to a cluster end, and the device includes: A device resource creation unit 610 is used to receive device information of a computing device deployed at a node and virtual device information of the node, and create device resources of the computing device based on the device information, wherein the device information is obtained by a device proxy component at the node; The binding resource creation unit 620 is used to create a device binding resource of the resource deployment unit corresponding to the resource management request based on the resource management request and the virtual device information when a resource management request is received; The resource management unit 630 is used to select a target device resource from all unbound device resources based on the device binding resources, and update the target device information in the device binding resources based on the device information corresponding to the target device resource, so that when the device proxy component monitors the resource management request, it can determine the target device information from the device binding resources of the resource deployment unit corresponding to the resource management request, and mount the computing device corresponding to the target device information to the resource deployment unit.

[0119] In the apparatus provided in an embodiment of the present invention, device information of a computing device deployed at a node is obtained and reported to a cluster end to create device resources. In addition, device binding resources are created on the cluster end based on a resource management request, thereby combining device resources and device binding resources to achieve fine-grained resource management and allocation, thereby improving the performance and efficiency of resource management in a heterogeneous cluster.

[0120] Based on any of the above embodiments, the device resources include device information, performance information and status information of the computing device.

[0121] Based on any of the above embodiments, the step of selecting a target device resource from all unbound device resources based on the device binding resource includes: Based on the status information of all device resources, determine the unbound device resources; Based on the device-bound resources, and the device information and performance information of the unbound device resources, a target device resource is selected from all unbound device resources.

[0122] Figure 7 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 7As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communication interface 720 and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute the resource management method, which includes: Acquire device information of a computing device deployed at a node, and report the device information to a cluster end, so as to trigger the cluster end to create device resources of the computing device based on the device information; Register the virtual device information of the node and report it to the cluster end; In the case of monitoring a resource management request, querying a resource deployment unit corresponding to the resource management request, determining target device information from device binding resources of the resource deployment unit, and mounting a computing device corresponding to the target device information to the resource deployment unit; The device binding resource is created by the cluster end based on the resource management request and the virtual device information, and the target device information is device information corresponding to the target device resource selected by the cluster end from all unbound device resources based on the device binding resource.

[0123] Alternatively, the method comprises: Receiving device information of a computing device deployed at a node and virtual device information of the node, and creating device resources of the computing device based on the device information, wherein the device information is obtained by a device proxy component at the node; In case of receiving a resource management request, creating a device binding resource of a resource deployment unit corresponding to the resource management request based on the resource management request and the virtual device information; Based on the device binding resources, a target device resource is selected from all unbound device resources, and the target device information in the device binding resources is updated based on the device information corresponding to the target device resource, so that when the device proxy component listens to the resource management request, it determines the target device information from the device binding resources of the resource deployment unit corresponding to the resource management request, and mounts the computing device corresponding to the target device information to the resource deployment unit.

[0124] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the relevant technology or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0125] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the resource management method provided by the above methods, the method includes: Acquire device information of a computing device deployed at a node, and report the device information to a cluster end, so as to trigger the cluster end to create device resources of the computing device based on the device information; Register the virtual device information of the node and report it to the cluster end; In the case of monitoring a resource management request, querying a resource deployment unit corresponding to the resource management request, determining target device information from device binding resources of the resource deployment unit, and mounting a computing device corresponding to the target device information to the resource deployment unit; The device binding resource is created by the cluster end based on the resource management request and the virtual device information, and the target device information is device information corresponding to the target device resource selected by the cluster end from all unbound device resources based on the device binding resource.

[0126] Alternatively, the method comprises: Receiving device information of a computing device deployed at a node and virtual device information of the node, and creating device resources of the computing device based on the device information, wherein the device information is obtained by a device proxy component at the node; In case of receiving a resource management request, creating a device binding resource of a resource deployment unit corresponding to the resource management request based on the resource management request and the virtual device information; Based on the device binding resources, a target device resource is selected from all unbound device resources, and the target device information in the device binding resources is updated based on the device information corresponding to the target device resource, so that when the device proxy component listens to the resource management request, it determines the target device information from the device binding resources of the resource deployment unit corresponding to the resource management request, and mounts the computing device corresponding to the target device information to the resource deployment unit.

[0127] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the resource management method provided by the above methods is implemented, and the method includes: Acquire device information of a computing device deployed at a node, and report the device information to a cluster end, so as to trigger the cluster end to create device resources of the computing device based on the device information; Register the virtual device information of the node and report it to the cluster end; In the case of monitoring a resource management request, querying a resource deployment unit corresponding to the resource management request, determining target device information from device binding resources of the resource deployment unit, and mounting a computing device corresponding to the target device information to the resource deployment unit; The device binding resource is created by the cluster end based on the resource management request and the virtual device information, and the target device information is device information corresponding to the target device resource selected by the cluster end from all unbound device resources based on the device binding resource.

[0128] Alternatively, the method comprises: Receiving device information of a computing device deployed at a node and virtual device information of the node, and creating device resources of the computing device based on the device information, wherein the device information is obtained by a device proxy component at the node; In case of receiving a resource management request, creating a device binding resource of a resource deployment unit corresponding to the resource management request based on the resource management request and the virtual device information; Based on the device binding resources, a target device resource is selected from all unbound device resources, and the target device information in the device binding resources is updated based on the device information corresponding to the target device resource, so that when the device proxy component listens to the resource management request, it determines the target device information from the device binding resources of the resource deployment unit corresponding to the resource management request, and mounts the computing device corresponding to the target device information to the resource deployment unit.

[0129] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0130] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A resource management method, characterized in that: include: Acquire device information of a computing device deployed at a node, and report the device information to a cluster end, so as to trigger the cluster end to create device resources of the computing device based on the device information; Register the virtual device information of the node and report it to the cluster end; In the case of monitoring a resource management request, querying a resource deployment unit corresponding to the resource management request, determining target device information from device binding resources of the resource deployment unit, and mounting a computing device corresponding to the target device information to the resource deployment unit; The device binding resource is created by the cluster end based on the resource management request and the virtual device information, and the target device information is device information corresponding to the target device resource selected by the cluster end from all unbound device resources based on the device binding resource; The registering the virtual device information of the node and reporting it to the cluster end includes: The node agent component of the node is based on registering virtual device information to trigger the node agent component to report the virtual device information to the cluster end, and the number of virtual devices included in the virtual device information is greater than or equal to the number of computing devices deployed at the node.

2. The resource management method according to claim 1, characterized in that: The obtaining of device information of a computing device deployed at a node includes: Scan the PCI device at the node to obtain the device model of the PCI device; Based on the device model of the PCI device, device information of a computing device corresponding to the PCI device is determined.

3. The resource management method according to claim 2, characterized in that: The determining, based on the device model of the PCI device, device information of the computing device corresponding to the PCI device includes: Determine, based on the device model of the PCI device, the device model of the computing device corresponding to the PCI device; Based on the correspondence between the general device model and the subdivided device model, the general device model corresponding to the device model of the computing device is determined, and the device information is determined based on the general device model.

4. The resource management method according to any one of claims 1 to 3, characterized in that: The device resources include device information, performance information, and status information of the computing device.

5. The resource management method according to any one of claims 1 to 3, characterized in that: The querying of the resource deployment unit corresponding to the resource management request includes: Receiving a virtual device allocation request sent by a node agent component of the node; Based on the virtual device identifier corresponding to the virtual device allocation request, the resource deployment unit corresponding to the resource management request is queried.

6. A resource management method, characterized in that: include: Receiving device information of a computing device deployed at a node and virtual device information of the node, and creating device resources of the computing device based on the device information, wherein the device information is obtained by a device agent component at the node, the virtual device information is registered and reported based on the node agent component of the node, and the number of virtual devices included in the virtual device information is greater than or equal to the number of computing devices deployed at the node; In case of receiving a resource management request, creating a device binding resource of a resource deployment unit corresponding to the resource management request based on the resource management request and the virtual device information; Based on the device binding resources, a target device resource is selected from all unbound device resources, and the target device information in the device binding resources is updated based on the device information corresponding to the target device resource, so that when the device proxy component listens to the resource management request, it determines the target device information from the device binding resources of the resource deployment unit corresponding to the resource management request, and mounts the computing device corresponding to the target device information to the resource deployment unit.

7. The resource management method according to claim 6, characterized in that: The device resources include device information, performance information, and status information of the computing device.

8. The resource management method according to claim 7, characterized in that: The selecting a target device resource from all unbound device resources based on the device binding resource includes: Based on the status information of all device resources, determine the unbound device resources; Based on the device-bound resources, and the device information and performance information of the unbound device resources, a target device resource is selected from all unbound device resources.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the resource management method according to any one of claims 1 to 8 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the resource management method according to any one of claims 1 to 8 is implemented.

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the resource management method according to any one of claims 1 to 8 is implemented.

Images