Computing resource allocation method, device, electronic device and computer readable medium

Abstract

The embodiments of the present disclosure disclose a computing power resource allocation method, device, electronic device and computer-readable medium. A specific implementation of the method includes: in response to detecting the login information of the target user, displaying the computing power device information interface and navigation bar; in response to detecting the target user's selection operation on the computing power resource allocation interface entry control, displaying the computing power resource allocation interface; in response to detecting the target user's selection operation on the computing power resource application control, according to the received computing power resource demand information, performing the following allocation steps: determining the user allocation computing power resource information according to the computing power resource demand information, user resource application authority information, user resource consumption score information and the remaining computing power resource information set; according to the user allocation computing power resource information, computing power resources are allocated to the target user to obtain the user computing power resource information. This implementation improves the utilization rate of computing power resources and user experience, and reduces the waste of computing power resources.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to a computing resource allocation method, device, electronic device, and computer-readable medium. Background Art

[0002] With the rapid development of deep learning technology, the demand for training large models continues to increase, which requires reasonable scheduling of computing power equipment to meet the training needs of large models. At present, the existing computing power platform mainly allocates computing power resources based on static priority and weight allocation strategies.

[0003] However, in practice, when the above method is used to allocate computing resources, the following technical problems often occur:

[0004] Static priority and weight allocation strategies for computing resource allocation are difficult to meet dynamically changing load and task requirements. In a multi-user, multi-task environment, there may be a situation where one user applies for more computing resources but fails to release the resources in time, while another user cannot apply for computing resources that meet their needs, resulting in uneven resource allocation, low utilization of computing resources, waste of computing resources and poor user experience.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention

[0006] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0007] Some embodiments of the present disclosure propose computing resource allocation methods, devices, electronic devices, and computer-readable media to solve one or more of the technical problems mentioned in the above background technology section.

[0008] In a first aspect, some embodiments of the present disclosure provide a computing power resource allocation method, the method comprising: in response to detecting the login information of a target user, displaying a computing power device information interface and a navigation bar; in response to detecting the target user's selection operation on the computing power resource details interface entry control in the navigation bar, displaying a computing power resource details interface; in response to detecting the target user's selection operation on the computing power resource allocation interface entry control in the navigation bar, displaying a computing power resource allocation interface, wherein the computing power resource allocation interface includes a computing power resource application control; in response to detecting the target user's selection operation on the computing power resource application control in the computing power resource allocation interface, displaying a computing power resource allocation interface according to the computing power resource application control submitted by the target user. The computing power resource demand information is used to perform the following allocation steps: obtaining the user resource consumption score information, user resource application authority information and remaining computing power resource information set of the above-mentioned target user, wherein the remaining computing power resource information in the above-mentioned remaining computing power resource information set includes the server identification information and each remaining computing power device information corresponding to the server identification information; determining the user allocated computing power resource information corresponding to the above-mentioned target user according to the above-mentioned computing power resource demand information, the above-mentioned user resource application authority information, the above-mentioned user resource consumption score information and the above-mentioned remaining computing power resource information set; allocating computing power resources to the above-mentioned target user according to the above-mentioned user allocated computing power resource information, and obtaining the user computing power resource information corresponding to the above-mentioned target user.

[0009] In a second aspect, some embodiments of the present disclosure provide a computing power resource allocation device, the device comprising: a first display unit, configured to display a computing power device information interface and a navigation bar in response to detecting login information of a target user; a second display unit, configured to display a computing power resource details interface in response to detecting a selection operation of the target user on an entry control of the computing power resource details interface in the navigation bar; a third display unit, configured to display a computing power resource allocation interface in response to detecting a selection operation of the target user on an entry control of the computing power resource allocation interface in the navigation bar, wherein the computing power resource allocation interface includes a computing power resource application control; and an allocation unit, configured to display a computing power resource application control in response to detecting a selection operation of the target user on an entry control of the computing power resource allocation interface Select an operation, and perform the following allocation steps according to the computing power resource demand information submitted by the above-mentioned target user: obtain the user resource consumption score information, user resource application authority information and remaining computing power resource information set of the above-mentioned target user, wherein the remaining computing power resource information in the above-mentioned remaining computing power resource information set includes the server identification information and each remaining computing power device information corresponding to the server identification information; determine the user allocation computing power resource information corresponding to the above-mentioned target user according to the above-mentioned computing power resource demand information, the above-mentioned user resource application authority information, the above-mentioned user resource consumption score information and the above-mentioned remaining computing power resource information set; allocate computing power resources to the above-mentioned target user according to the above-mentioned user allocation computing power resource information, and obtain the user computing power resource information corresponding to the above-mentioned target user.

[0010] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the above-mentioned first aspect.

[0011] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method described in any implementation manner of the above-mentioned first aspect is implemented.

[0012] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the computing power resource allocation method of some embodiments of the present disclosure, the utilization rate of computing power resources and user experience are improved, and the waste of computing power resources is reduced. Specifically, the reasons for the low utilization rate and user experience of computing power resources and the high waste of computing power resources are: the static priority and weight allocation strategy for computing power resource allocation is difficult to meet the dynamically changing load and task requirements. When in a multi-user, multi-task environment, there may be a user who applies for more computing power resources but fails to release resources in time, while another user cannot apply for computing power resources that meet the requirements, resulting in uneven resource allocation, resulting in low utilization of computing power resources, resulting in waste of computing power resources and poor user experience. Based on this, the computing power resource allocation method of some embodiments of the present disclosure, first, in response to detecting the login information of the target user, displays the computing power device information interface and navigation bar. Thus, the computing power device information interface and navigation bar can be displayed. Users can view the allocation information of each computing power device in the computing power device information interface. Then, in response to detecting the target user's selection operation on the computing power resource details interface entry control in the navigation bar, the computing power resource details interface is displayed. Thus, the computing power resource details interface can be displayed. The user can view the GPU usage details so that the user can formulate a computing power resource application plan. Afterwards, in response to detecting the target user's selection operation on the computing power resource allocation interface entry control in the navigation bar, the computing power resource allocation interface is displayed, wherein the computing power resource allocation interface includes a computing power resource application control. Thus, the computing power resource allocation interface can be displayed. The user can view the information of each allocated computing power resource and apply for computing power resources for the user. Next, in response to detecting the target user's selection operation on the computing power resource application control in the computing power resource allocation interface, the following allocation steps are performed according to the computing power resource demand information submitted by the target user received: First, the user resource consumption score information, user resource application authority information and remaining computing power resource information set of the target user are obtained, wherein the remaining computing power resource information in the remaining computing power resource information set includes server identification information and each remaining computing power device information corresponding to the server identification information. Thus, the user resource consumption score information, user resource application authority information and remaining computing power resource information set of the target user can be obtained. Then, based on the computing power resource demand information, the user resource application authority information, the user resource consumption score information and the remaining computing power resource information set, the user allocated computing power resource information corresponding to the target user is determined. Thus, the user allocated computing power resource information corresponding to the target user can be obtained. Among them, the user allocated computing power resource information can represent the computing power resource information and allocation time information to be allocated to the target user.Finally, according to the above-mentioned user allocation computing power resource information, computing power resources are allocated to the above-mentioned target user to obtain the user computing power resource information corresponding to the above-mentioned target user. In this way, the user computing power resource information corresponding to the above-mentioned target user can be obtained. The user can use the computing power resources corresponding to the above-mentioned user computing power resource information to perform tasks such as model training. Also because the implemented computing power resource allocation method can support and limit the user's application for computing power resources according to the user's resource consumption score information, when the user's user resource application authority is insufficient to apply for the required computing power resources, and the remaining computing power resources meet the computing power resources required by the user, this part of the resources will be occupied with the preempted authority, avoiding the situation where there are idle computing power resources and the user cannot apply for resources. When the user with sufficient resource consumption score information and satisfied user resource application authority applies for computing power resources, and the remaining computing power resources do not meet the computing power resources required by the user, and the computing power resources required by the user are met in the preempted resources, the computing power resources in the preempted resources that meet the user's requirements will be forcibly allocated to the user. It not only ensures that users with insufficient resource application permissions can apply for preempted resources, but also ensures that users with sufficient resource consumption score information and user resource application permissions can apply for the required computing resources in a timely manner when the remaining computing resources are insufficient. This improves the utilization rate of computing resources, reduces the waste of computing resources, and improves the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0014] Figure 1 is a flow chart of some embodiments of the computing power resource allocation method according to the present disclosure;

[0015] Figure 2 It is a schematic diagram of the structure of some embodiments of the computing power resource allocation device according to the present disclosure;

[0016] Figure 3 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0017] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0018] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0019] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0020] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0021] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0022] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0023] Figure 1 The process 100 of some embodiments of the computing resource allocation method according to the present disclosure is shown. The computing resource allocation method includes the following steps:

[0024] Step 101, in response to detecting the login information of the target user, displaying the computing power device information interface and the navigation bar.

[0025] In some embodiments, in response to detecting the login information of the target user, the execution subject (e.g., computing device) of the computing power resource information allocation method may display a computing power device information interface and a navigation bar. The above login information may be the identity information submitted by the above target user for login. The above identity information may include, but is not limited to: user ID, user login password. The above computing power device information interface displays the allocation information of each computing power device. The computing power device allocation information in the above each computing power device allocation information may be the relevant information of the allocated computing power device. The above computing power device allocation information may include, but is not limited to, multiple items of the following: device name, device IP address, graphics card model, graphics card driver version number, number of allocated GPUs, and total number of GPUs. The above navigation bar may include, but is not limited to, multiple items of the following: computing power device information interface entry control, computing power resource details interface entry control, and computing power resource allocation interface entry control.

[0026] Step 102, in response to detecting a target user's selection operation on an entry control of a computing power resource details interface in a navigation bar, display a computing power resource details interface.

[0027] In some embodiments, in response to detecting the target user's selection operation on the computing power resource details interface entry control in the navigation bar, the execution subject may display the computing power resource details interface. The computing power resource details interface entry control may be a selection control for entering the computing power resource details interface. The computing power resource details interface may be an interface for displaying GPU usage details. The computing power resource details interface may display various computing power resource details selection controls. The various computing power resource details selection controls may be various selection controls for viewing various computing power resource details information. In some embodiments, in response to detecting the selection operation on the computing power resource details selection control in the various computing power resource details selection controls, the computing power resource details information corresponding to the selected computing power resource details selection control may be displayed in the computing power resource details interface. The computing power resource details information may be the GPU usage of the target server counted at a preset statistical time interval. The target server may be the server corresponding to the computing power resource details selection control. Here, the specific setting of the preset statistical time interval is not limited. The computing power resource details information may include but is not limited to multiple items of the following: server identification information, total number of GPUs, number of allocated GPUs, number of allocated and occupied GPUs, and number of GPUs in use. The above server identification information can represent the GPU configuration of the server. Therefore, the user can formulate a suitable computing resource application plan based on the detailed information of the GPU usage of the server.

[0028] Step 103, in response to detecting a selection operation of the target user on the computing power resource allocation interface entry control in the navigation bar, displaying the computing power resource allocation interface.

[0029] In some embodiments, in response to detecting the target user's selection operation on the computing power resource allocation interface entry control in the navigation bar, the execution entity may display the computing power resource allocation interface. Among them, the computing power resource allocation interface may be an interface for users to view their own allocated computing power resource information and for users to apply for computing power resources. The computing power resource allocation interface may include a computing power resource application control. The computing power resource application control may be a selection control for applying for computing power resources. The allocated computing power resource information in the above-mentioned each allocated computing power resource information may be information related to computing power resources allocated to the current user. The above-mentioned allocated computing power resource information may include, but is not limited to, multiple items of the following: device name, graphics card model, number of GPUs, allocation time, running time, device IP address, and port number.

[0030] Step 104, in response to detecting the target user's selection operation on the computing resource application control in the computing resource allocation interface, the following allocation steps are performed according to the computing resource demand information received and submitted by the target user:

[0031] Step 1041, obtaining the target user's user resource consumption score information, user resource application authority information and remaining computing power resource information set.

[0032] In some embodiments, the execution subject may obtain the user resource consumption score information, user resource application authority information and remaining computing power resource information set of the target user. Among them, the computing power resource demand information may be request information for applying for computing power resources. The computing power resources may be GPU resources. The user resource application authority information may be the number of GPU devices corresponding to each type of server that the user can apply for at most. Each type of server in the above types of servers corresponds to the number of GPU devices in the above numbers of GPU devices. The user resource application authority information may include the computing power device quantity information corresponding to each server identification information and each server identification information in the above server identification information. The computing power device quantity information may be the number of GPU devices. The remaining computing power resource information set may represent the computing power resource information for which resources can be allocated. The computing power resource information in the above computing power resource information may represent the server identification information and the number of GPU devices corresponding to the server identification information. The server identification information may be, but is not limited to, one of the following: A800, RTX3090, A100, RTX4090. The remaining computing power resource information in the above-mentioned remaining computing power resource information set may include server identification information and each remaining computing power device information corresponding to the server identification information. In practice, the above-mentioned execution subject can obtain the user resource consumption score information and user resource application permission information of the above-mentioned target user from the associated storage device. Each target server can be monitored by the zabbix monitoring tool to obtain the remaining computing power resource information set. It should be noted that each of the above-mentioned target servers can be managed by the Kubernetes tool and InfiniBand (IB) network technology.

[0033] Step 1042, determine the user allocated computing power resource information corresponding to the target user based on the computing power resource demand information, user resource application authority information, user resource consumption score information and the remaining computing power resource information set.

[0034] In some embodiments, the execution subject may determine the user allocation computing power resource information corresponding to the target user according to the computing power resource demand information, the user resource application authority information, the user resource consumption score information and the remaining computing power resource information set. The computing power resource demand information may include server identification information, computing power device quantity information and application time information. The user resource consumption score information may be the score used by the user to apply for computing power resources. For example, the user resource consumption score information may be points. The computing power device quantity information may be the number of GPUs applied for by the user. The application time information may be the duration of the application for the use of computing power devices. The computing power device may be a GPU. The user allocation computing power resource information may represent the computing power resource information and allocation time information to be allocated to the target user. The allocation time information may represent the time for the computing power resources to be allocated to the target user. In practice, the execution subject may determine the user allocation computing power resource information corresponding to the target user according to the computing power resource demand information, the user resource application authority information, the user resource consumption score information and the remaining computing power resource information set in various ways.

[0035] In some optional implementations of some embodiments, the execution subject may determine the user allocated computing power resource information corresponding to the target user according to the computing power resource demand information, the user resource application authority information, the user resource consumption score information and the remaining computing power resource information set through the following steps:

[0036] The first step is to determine whether the computing power resource demand information meets the preset application conditions based on the user resource application permission information. The preset application condition may be that the user resource application permission information is greater than or equal to the computing power device quantity information included in the computing power resource demand information. In practice, in response to determining that the user resource application permission information is greater than or equal to the computing power device quantity information included in the computing power resource demand information, the execution subject may determine that the computing power resource demand information meets the preset application conditions. In response to determining that the user resource application permission information is less than the computing power device quantity information included in the computing power resource demand information, the execution subject may determine that the computing power resource demand information does not meet the preset application conditions.

[0037] The second step is to determine whether the above-mentioned remaining computing power resource information set meets the preset application conditions in response to determining that the above-mentioned computing power resource demand information meets the above-mentioned preset application conditions. Among them, the above-mentioned preset allocation condition can be that the number of remaining computing power devices corresponding to the required server identification information in the above-mentioned remaining computing power resource information set is greater than or equal to the number of required computing power devices. The above-mentioned required server identification information can be the server identification information included in the above-mentioned computing power resource demand information. The above-mentioned number of required computing power devices can be the computing power device number information corresponding to the above-mentioned required server identification information included in the above-mentioned computing power resource demand information. In practice, in response to determining that the number of remaining computing power devices corresponding to the required server identification information in the above-mentioned remaining computing power resource information set is greater than or equal to the number of required computing power devices, the above-mentioned execution subject can determine that the above-mentioned remaining computing power resource information set meets the above-mentioned preset allocation conditions. In response to determining that the number of remaining computing power devices corresponding to the required server identification information in the above-mentioned remaining computing power resource information set is less than the number of required computing power devices, the above-mentioned execution subject can determine that the above-mentioned remaining computing power resource information set does not meet the above-mentioned preset allocation conditions.

[0038] In the third step, in response to determining that the above-mentioned remaining computing power resource information set meets the above-mentioned preset allocation condition, the following steps are performed:

[0039] The first sub-step is to determine the current time information as the allocation time information.

[0040] The second sub-step is to update the above-mentioned remaining computing power resource information set according to the above-mentioned computing power resource demand information. In practice, the above-mentioned execution entity may subtract the number of the above-mentioned required computing power devices from the number of the remaining computing power devices corresponding to the above-mentioned required server identification information in the above-mentioned remaining computing power resource information set to update the above-mentioned remaining computing power resource information set.

[0041] The third sub-step is to determine the server identification information and computing power equipment quantity information included in the above-mentioned allocation time information and the above-mentioned computing power resource demand information as the user allocation computing power resource information corresponding to the above-mentioned target user.

[0042] Optionally, the above execution entity may further perform the following steps:

[0043] The first step is to determine, in response to determining that the computing power resource demand information does not meet the preset application conditions, the pre-allocated computing power resource demand information and the computing power resource demand information to be preempted are determined according to the user resource application authority information and the computing power resource demand information. Among them, the pre-allocated computing power resource demand information can be the number of GPUs corresponding to the demand server identification information that the target user can currently apply for. The computing power resource demand information to be preempted can represent the number of GPUs corresponding to the demand server identification information in the computing power resource demand information that exceeds the user resource application authority information. In practice, first, the execution subject can determine the computing power device quantity information corresponding to the demand server identification information in the user resource application authority information as the pre-allocated computing power device quantity information. Then, the demand server identification information and the pre-allocated computing power device quantity information can be determined as the pre-allocated computing power resource demand information. Afterwards, the difference between the demand computing power device quantity and the pre-allocated computing power device quantity information can be determined as the computing power device quantity information to be preempted. Finally, the demand server identification information and the computing power device quantity information to be preempted can be determined as the computing power resource demand information to be preempted.

[0044] In the second step, the current time information is determined as the allocation time information.

[0045] The third step is to update the remaining computing power resource information set according to the above-mentioned pre-allocated computing power resource demand information. In practice, the above-mentioned execution subject can subtract the pre-allocated computing power device quantity information in the above-mentioned pre-allocated computing power resource demand information from the remaining computing power device quantity information corresponding to the above-mentioned demand server identification information in the above-mentioned remaining computing power resource information set to update the remaining computing power resource information set.

[0046] The fourth step is to determine the allocation time information, the server identification information and the computing power device quantity information included in the above-mentioned pre-allocated computing power resource demand information as the non-preemptive computing power resource information corresponding to the above-mentioned target user.

[0047] The fifth step is to determine whether the remaining computing power resource information set meets the preset conditions for preempting computing power resources based on the above-mentioned computing power resource demand information to be preempted. Among them, the above-mentioned preset conditions for preempting computing power resources can be that the remaining computing power device quantity information corresponding to the above-mentioned demand server identification information in the above-mentioned remaining computing power resource information set is greater than or equal to the number of computing power devices to be preempted in the above-mentioned computing power resource demand information to be preempted. In practice, in response to determining that the remaining computing power device quantity information corresponding to the above-mentioned demand server identification information in the above-mentioned remaining computing power resource information set is greater than or equal to the number of computing power devices to be preempted in the above-mentioned computing power resource demand information to be preempted, the above-mentioned execution subject can determine that the remaining computing power resource information meets the preset conditions for preempting computing power resources. In response to determining that the remaining computing power device quantity information corresponding to the above-mentioned demand server identification information in the above-mentioned remaining computing power resource information set is less than the number of computing power devices to be preempted in the above-mentioned computing power resource demand information to be preempted, the above-mentioned execution subject can determine that the remaining computing power resource information does not meet the above-mentioned preset conditions for preempting computing power resources.

[0048] Step 6: In response to determining that the remaining computing power resource information set meets the above-mentioned preset preempting computing power resource conditions, the remaining computing power resource information set is updated according to the above-mentioned computing power resource demand information to be preempted. In practice, the above-mentioned execution subject can subtract the remaining computing power device quantity information corresponding to the above-mentioned demand server identification information from the above-mentioned computing power resource demand information to be preempted, so as to update the remaining computing power resource information set.

[0049] The seventh step is to determine the allocation time information, the server identification information included in the above-mentioned computing power resource demand information to be preempted, and the computing power device quantity information as the preempted computing power resource information corresponding to the above-mentioned target user.

[0050] In the eighth step, the server identification information and computing power equipment quantity information included in the computing power resource demand information to be preempted are added as preempted equipment information to the preset preempted equipment information set. Among them, the preset preempted equipment information in the preset preempted equipment information set can be the computing power resource information whose usage permission is the preempted permission. It should be noted that the usage permission of the computing power equipment can be the preempted permission or the non-preempted permission. The priority of the non-preempted permission can be greater than the priority of the preempted permission.

[0051] In a ninth step, the non-preemptive computing power resource information and the preemptive computing power resource information are determined as the user allocated computing power resource information corresponding to the target user.

[0052] Optionally, the above execution entity may further perform the following steps:

[0053] The first step is to determine whether there is preset preemptive device information corresponding to the computing resource demand information in the preset preemptive device information set in response to determining that the above-mentioned remaining computing power resource information set does not meet the above-mentioned preset allocation conditions. In practice, in response to determining that there is preset preemptive device information in the preset preemptive device information set whose server identification information is the above-mentioned required server identification information and whose computing power device quantity information is greater than or equal to the above-mentioned required computing power device quantity, it is determined that there is preemptive device information corresponding to the above-mentioned computing power resource demand information in the preset preemptive device information set.

[0054] In the second step, in response to determining that the preset preemption device information set contains preset preemption device information corresponding to the computing power resource demand information, the following steps are performed:

[0055] In a first sub-step, any one of the preset preemptive device information corresponding to the computing power resource demand information in the preset preemptive device information set is determined as the pre-allocated device information.

[0056] The second sub-step is to back up the server corresponding to the pre-allocated device information. In practice, the execution subject may use the etcd tool to back up the server corresponding to the pre-allocated device information. It should be noted that the execution subject may use the CephFS distributed storage tool to store the backup data.

[0057] The third sub-step is to close each task running in the server corresponding to the pre-allocated device information.

[0058] The fourth sub-step is to delete the above-mentioned pre-allocated device information from the preset preempted device information set.

[0059] The fifth sub-step is to determine the current time information as the allocation time information.

[0060] The sixth sub-step is to determine the above-mentioned pre-allocated device information and allocation time information as the user allocation computing power resource information corresponding to the above-mentioned target user.

[0061] Step 1043, allocate computing resources to the target user according to the user allocation computing resource information, and obtain the user computing resource information corresponding to the target user.

[0062] In some embodiments, the execution subject may allocate computing resources to the target user according to the user allocation computing resource information, and obtain user computing resource information corresponding to the target user. The user computing resource information may represent the computing resources allocated to the target user. In practice, the execution subject may allocate computing resources to the target user according to the user allocation computing resource information through Kubernetes resource scheduling technology, and obtain user computing resource information corresponding to the target user.

[0063] Optionally, the above execution entity may further perform the following steps:

[0064] The first step is to determine the value attribute information of the server corresponding to the above-mentioned server identification information based on the server identification information included in the above-mentioned user computing power resource information. Among them, the above-mentioned value attribute information can represent the value of the server corresponding to the above-mentioned server identification information. For example, the above-mentioned value attribute information can be a unit price. In practice, the above-mentioned execution entity can determine the value attribute information included in the preset server identification and value relationship information set corresponding to the above-mentioned server identification information as the value attribute information of the server corresponding to the above-mentioned server identification information. The above-mentioned preset server identification and value relationship information set can represent the correspondence between the server identification information and the value attribute information. The preset server identification and value relationship information in the above-mentioned preset server identification and value relationship information set can include the server identification information and the value attribute information corresponding to the server identification information.

[0065] The second step is to determine the allocation time information included in the above user computing power resource information as the target time information.

[0066] The third step is to monitor and process the above-mentioned user computing power resource information based on the above-mentioned target time information and the above-mentioned value attribute information.

[0067] Next, when the above-mentioned user computing power resource information is monitored and processed by the present disclosure, the following technical problem 2 further exists:

[0068] Users usually apply for multiple computing devices, but fail to monitor the ratio of idle devices to the total number of computing devices applied for within a certain period of time and impose corresponding penalties. This leads to a large number of users applying for arbitrary numbers of computing devices, resulting in a waste of computing resources.

[0069] In some optional implementations of some embodiments, the execution subject may monitor and process the user computing power resource information based on the target time information and the value attribute information through the following steps:

[0070] The first step is to perform the following monitoring steps based on the target time information:

[0071] The first monitoring step is to determine whether the current time information satisfies a preset time interval condition according to the target time information. The preset time interval condition may be that the time interval between the current time information and the target time information is equal to the preset time interval. Here, the specific setting of the preset time interval is not limited. For example, the preset time interval may be 20 minutes.

[0072] The second monitoring step is to determine the idle time information of each computing device resource included in the user computing resource in response to determining that the current time information satisfies the preset time interval condition. Among them, the idle time information in the above idle time information can be the duration of continuous idleness of the computing device. In practice, the above execution subject can use Prometheus and GPU Exporter tools to determine the idle time information of each computing device corresponding to each computing device information included in the user computing resource information.

[0073] The third monitoring step is to determine whether there is computing power equipment resource information that meets the preset idle condition in the above-mentioned user computing power resource information based on the above-mentioned idle time information. Among them, the above-mentioned preset idle condition can be that the idle time information is greater than the preset idle time. Here, the specific setting of the above-mentioned preset idle time is not limited. For example, the preset idle time can be 20 minutes.

[0074] In a fourth monitoring step, in response to determining that computing power equipment resource information that meets the preset idle condition exists in the user computing power resource information, each computing power equipment resource information that meets the preset idle condition in the user computing power resource information is determined as each idle computing power equipment resource information. The above-mentioned each idle computing power equipment information may represent each computing power equipment whose idle time is greater than the above-mentioned preset idle time information.

[0075] The fifth monitoring step is to generate resource consumption score deduction ratio information based on the number of each computing power device resource information included in the above user computing power resource information and the number of each idle computing power device resource information. The above resource consumption score deduction ratio information may be the ratio of the user resource consumption score of the above target user to be deducted. In practice, the above execution entity may determine the ratio of the number of each idle computing power device information to the number of each computing power device information as the resource consumption score deduction ratio information.

[0076] The sixth monitoring step is to generate deduction score information according to the resource consumption score deduction ratio information and the value attribute information. The deduction score information may be the user resource consumption score information consumed by each GPU applied by the user within the preset time interval. In practice, first, the execution subject may determine the deduction score corresponding to the value attribute information in the preset device value and deduction score relationship set as the basic deduction score. Then, the sum of the basic deduction score and the resource consumption score deduction ratio information may be determined as the deduction ratio information. The deduction ratio information may represent the user resource consumption score information to be deducted per minute by a single GPU. After that, the product of the deduction ratio information and the number of each computing power device information included in the user computing power resource information may be determined as the unit deduction score. Finally, the product of the unit deduction score and the preset value may be determined as the deduction score information. The preset device value and deduction score relationship set may represent the corresponding relationship between the value attribute information and the deduction score. The preset device value and deduction score relationship in the preset device value and deduction score relationship set may include value attribute information and the deduction score corresponding to the value attribute information. The above-mentioned unit deduction score may be the user resource consumption score information consumed per minute by each GPU applied by the user.

[0077] The seventh monitoring step is to update the user resource consumption score information of the target user according to the deduction score information. In practice, the execution subject may subtract the deduction score information from the user resource consumption score information to update the user resource consumption score information.

[0078] In the eighth monitoring step, in response to determining that there is computing power equipment resource information that meets the preset warning condition in each computing power equipment resource information included in the above-mentioned user computing power resource information, the computing power equipment idle warning information is sent to the associated terminal device. Among them, the above-mentioned terminal device can be but not limited to: a mobile phone, a computer. The above-mentioned preset warning condition can be that the remaining use time of the computing power equipment corresponding to the computing power equipment information is less than the preset time. Here, the specific setting of the above-mentioned preset time is not limited. For example, the preset time can be 20 minutes. The above-mentioned computing power equipment idle warning information can be a prompt information that reminds the user that the computing power equipment is about to reach the use limit time. The above-mentioned remaining use time can be the time when the computing power equipment can continue to be used. The above-mentioned remaining use time can be the time interval between the accumulated use time and the application time. The above-mentioned application time can be the application time included in the above-mentioned computing power resource demand information. In practice, in response to determining that there is computing power equipment information whose remaining use time is less than the above-mentioned preset time in the above-mentioned computing power equipment information, the above-mentioned execution subject can send the computing power equipment idle warning information to the associated terminal device in the form of an email or a text message.

[0079] The ninth monitoring step, in response to determining that there is computing power device resource information that meets the preset shutdown condition in each computing power device resource information included in the above-mentioned user computing power resource information, execute the following steps:

[0080] The first sub-step is to send a computing device shutdown prompt message to the above-mentioned associated terminal device, wherein the computing device shutdown prompt message may be a prompt message to prompt the user that the computing device is shut down.

[0081] The second sub-step is to recycle the computing power equipment resources corresponding to the above-mentioned idle computing power equipment resource information. In practice, the above-mentioned execution subject can delete the above-mentioned idle computing power equipment resource information from the above-mentioned user computing power resource information to recycle the computing power equipment resources corresponding to the above-mentioned idle computing power equipment resource information.

[0082] The third sub-step is to determine the above-mentioned idle computing power equipment resource information and the server identification information corresponding to the above-mentioned idle computing power equipment resource information as the remaining computing power resource information.

[0083] The fourth sub-step is to add the above remaining computing power resource information to the remaining computing power resource information set to update the remaining computing power resource information set.

[0084] The tenth monitoring step is to use the current time information as the target time information and execute the above monitoring steps again based on the target time information.

[0085] The first monitoring step to the tenth monitoring step of the above technical solution and their related contents are an inventive point of the embodiment of the present disclosure, which solves the above technical problem 2 "Users usually apply for multiple computing devices, and fail to monitor and punish the proportion of the number of idle devices in a certain period of time to the total number of computing devices applied for, resulting in a large number of users applying for arbitrary number of computing devices, thereby resulting in a waste of computing resources". The factors that lead to the waste of computing resources are often as follows: Users usually apply for multiple computing devices, and fail to monitor and punish the proportion of the number of idle devices in a certain period of time to the total number of computing devices applied for, resulting in a large number of users applying for arbitrary number of computing devices. If the above factors are solved, the effect of reducing the waste of computing resources can be achieved. In order to achieve this effect, the present disclosure introduces monitoring and punishing the proportion of the number of idle devices in a certain period of time to the total number of computing devices applied for, and deducting user points in units of minutes according to the proportion of the number of idle devices to the total number of computing devices applied for, thereby encouraging users to apply for computing resources more reasonably, and also reducing the situation where the same user applies for a large number of computing resources at the same time, thereby reducing the waste of computing resources.

[0086] Optionally, first, in response to determining that the target user is a first-level administrator user and detecting that the target user performs an allocation operation on the user resource consumption score information, the following allocation steps are performed:

[0087] The first step is to generate an allocation task corresponding to the user identification information and the preset score information in response to detecting the user identification information and the preset score information submitted by the target user. Among them, the first-level administrator user can be a user who has the authority to allocate points to the second-level administrator user. The user identification information can be a user ID. The second-level administrator user can be a user who has the authority to allocate points to ordinary users. For example, the second-level administrator user can be a tutor user. Ordinary users can be student users. In practice, the execution entity can determine the preset allocation task as the allocation task corresponding to the user identification information and the preset score information. The preset allocation task can be a timed task that allocates the preset score information to the user corresponding to the user identification information at a preset time. Here, the specific setting of the preset time is not limited.

[0088] The second step is to add the above-mentioned assigned tasks to an assigned task queue, wherein the above-mentioned assigned task queue can represent each assigned task to be executed.

[0089] Then, in response to determining that the current time meets the preset allocation time condition, each allocation task in the obtained allocation task queue is executed. The preset allocation time condition may be that the current time is the time for executing each allocation task.

[0090] Optionally, in response to determining that the target user is a secondary administrator user and detecting that the target user performs an allocation operation on the user resource consumption score information, the following allocation steps are performed:

[0091] The first step is to update the user resource consumption score information of the target user according to the user identification information and pre-allocated score information submitted by the target user. In practice, the execution entity may subtract the pre-allocated score information from the user resource consumption score information to update the user resource consumption score information of the target user.

[0092] The second step is to determine the pre-allocated score information as the user resource consumption score information corresponding to the user identification information.

[0093] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the computing power resource allocation method of some embodiments of the present disclosure, the utilization rate of computing power resources and user experience are improved, and the waste of computing power resources is reduced. Specifically, the reasons for the low utilization rate and user experience of computing power resources and the high waste of computing power resources are: the static priority and weight allocation strategy for computing power resource allocation is difficult to meet the dynamically changing load and task requirements. When in a multi-user, multi-task environment, there may be a user who applies for more computing power resources but fails to release resources in time, while another user cannot apply for computing power resources that meet the requirements, resulting in uneven resource allocation, resulting in low utilization of computing power resources, resulting in waste of computing power resources and poor user experience. Based on this, the computing power resource allocation method of some embodiments of the present disclosure, first, in response to detecting the login information of the target user, displays the computing power device information interface and navigation bar. Thus, the computing power device information interface and navigation bar can be displayed. Users can view the allocation information of each computing power device in the computing power device information interface. Then, in response to detecting the target user's selection operation on the computing power resource details interface entry control in the navigation bar, the computing power resource details interface is displayed. Thus, the computing power resource details interface can be displayed. The user can view the GPU usage details so that the user can formulate a computing power resource application plan. Afterwards, in response to detecting the target user's selection operation on the computing power resource allocation interface entry control in the navigation bar, the computing power resource allocation interface is displayed, wherein the computing power resource allocation interface includes a computing power resource application control. Thus, the computing power resource allocation interface can be displayed. The user can view the information of each allocated computing power resource and apply for computing power resources for the user. Next, in response to detecting the target user's selection operation on the computing power resource application control in the computing power resource allocation interface, the following allocation steps are performed according to the computing power resource demand information submitted by the target user received: First, the user resource consumption score information, user resource application authority information and remaining computing power resource information set of the target user are obtained, wherein the remaining computing power resource information in the remaining computing power resource information set includes server identification information and each remaining computing power device information corresponding to the server identification information. Thus, the user resource consumption score information, user resource application authority information and remaining computing power resource information set of the target user can be obtained. Then, based on the computing power resource demand information, the user resource application authority information, the user resource consumption score information and the remaining computing power resource information set, the user allocated computing power resource information corresponding to the target user is determined. Thus, the user allocated computing power resource information corresponding to the target user can be obtained. Among them, the user allocated computing power resource information can represent the computing power resource information and allocation time information to be allocated to the target user.Finally, according to the above-mentioned user allocation computing power resource information, computing power resources are allocated to the above-mentioned target user to obtain the user computing power resource information corresponding to the above-mentioned target user. In this way, the user computing power resource information corresponding to the above-mentioned target user can be obtained. The user can use the computing power resources corresponding to the above-mentioned user computing power resource information to perform tasks such as model training. Also because the implemented computing power resource allocation method can support and limit the user's application for computing power resources according to the user's resource consumption score information, when the user's user resource application authority is insufficient to apply for the required computing power resources, and the remaining computing power resources meet the computing power resources required by the user, this part of the resources will be occupied with the preempted authority, avoiding the situation where there are idle computing power resources and the user cannot apply for resources. When the user with sufficient resource consumption score information and satisfied user resource application authority applies for computing power resources, and the remaining computing power resources do not meet the computing power resources required by the user, and the computing power resources required by the user are met in the preempted resources, the computing power resources in the preempted resources that meet the user's requirements will be forcibly allocated to the user. It not only ensures that users with insufficient resource application permissions can apply for preempted resources, but also ensures that users with sufficient resource consumption score information and user resource application permissions can apply for the required computing resources in a timely manner when the remaining computing resources are insufficient. This improves the utilization rate of computing resources, reduces the waste of computing resources, and improves the user experience.

[0094] Further references Figure 2 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a computing resource allocation device. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0095] like Figure 2As shown, the computing power resource allocation device 200 of some embodiments includes: a first display unit 201, a second display unit 202, a third display unit 203 and an allocation unit 204. Among them, the first display unit 201 is configured to display a computing power device information interface and a navigation bar in response to detecting the login information of the target user; the second display unit 202 is configured to display a computing power resource details interface in response to detecting the selection operation of the above-mentioned target user on the computing power resource details interface entry control in the above-mentioned navigation bar; the third display unit 203 is configured to display a computing power resource allocation interface in response to detecting the selection operation of the above-mentioned target user on the computing power resource allocation interface entry control in the above-mentioned navigation bar, wherein the above-mentioned computing power resource allocation interface includes a computing power resource application control; the allocation unit 204 is configured to display a computing power resource application control in response to detecting the selection operation of the above-mentioned target user on the computing power resource application control in the above-mentioned computing power resource allocation interface, according to the received The computing power resource demand information submitted by the above-mentioned target user is used to perform the following allocation steps: obtaining the user resource consumption score information, user resource application authority information and remaining computing power resource information set of the above-mentioned target user, wherein the remaining computing power resource information in the above-mentioned remaining computing power resource information set includes the server identification information and each remaining computing power device information corresponding to the server identification information; determining the user allocated computing power resource information corresponding to the above-mentioned target user based on the above-mentioned computing power resource demand information, the above-mentioned user resource application authority information, the above-mentioned user resource consumption score information and the above-mentioned remaining computing power resource information set; and allocating computing power resources to the above-mentioned target user based on the above-mentioned user allocated computing power resource information to obtain the user computing power resource information corresponding to the above-mentioned target user.

[0096] It is understandable that the units recorded in the computing resource allocation device 200 and the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the device 200 and the units included therein, and will not be described in detail here.

[0097] Reference below Figure 3 , which shows a structural schematic diagram of an electronic device (eg, a computing device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0098] like Figure 3As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0099] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0100] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0101] It should be noted that the computer-readable medium recorded in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0102] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0103] The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or it may exist independently without being installed in the electronic device. The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device: in response to detecting the login information of the target user, displays the computing power device information interface and the navigation bar; in response to detecting the selection operation of the above-mentioned target user on the computing power resource details interface entry control in the above-mentioned navigation bar, displays the computing power resource details interface; in response to detecting the selection operation of the above-mentioned target user on the computing power resource allocation interface entry control in the above-mentioned navigation bar, displays the computing power resource allocation interface, wherein the above-mentioned computing power resource allocation interface includes a computing power resource application control; in response to detecting the selection operation of the above-mentioned target user on the computing power resource application control in the above-mentioned computing power resource allocation interface, according to the received above-mentioned The computing power resource demand information submitted by the target user is used to perform the following allocation steps: obtaining the user resource consumption score information, user resource application authority information and a set of remaining computing power resource information of the above target user, wherein the remaining computing power resource information in the above remaining computing power resource information set includes the server identification information and each remaining computing power device information corresponding to the server identification information; determining the user allocated computing power resource information corresponding to the above target user based on the above computing power resource demand information, the above user resource application authority information, the above user resource consumption score information and the above remaining computing power resource information set; and allocating computing power resources to the above target user based on the above user allocated computing power resource information to obtain the user computing power resource information corresponding to the above target user.

[0104] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0105] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0106] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The described units may also be provided in a processor, for example, may be described as: a processor comprising a first display unit, a second display unit, a third display unit and an allocation unit. The names of these units do not, in some cases, constitute limitations on the units themselves, for example, the first display unit may also be described as "a unit that displays a computing power device information interface and a navigation bar in response to detecting the login information of a target user".

[0107] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0108] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.

Claims

1. A computing resource allocation method, comprising: In response to detecting the login information of the target user, displaying a computing power device information interface and a navigation bar; In response to detecting a selection operation of the target user on an entry control of the computing power resource details interface in the navigation bar, displaying a computing power resource details interface; In response to detecting a selection operation of the target user on an entry control of a computing power resource allocation interface in the navigation bar, displaying a computing power resource allocation interface, wherein the computing power resource allocation interface includes a computing power resource application control; In response to detecting a selection operation of the target user on a computing resource application control in the computing resource allocation interface, the following allocation steps are performed according to the computing resource demand information received and submitted by the target user: Obtain user resource consumption score information, user resource application authority information, and a remaining computing power resource information set of the target user, wherein the remaining computing power resource information in the remaining computing power resource information set includes server identification information and each remaining computing power device information corresponding to the server identification information; Determine user allocation computing resource information corresponding to the target user according to the computing resource demand information, the user resource application authority information, the user resource consumption score information and the remaining computing resource information set; Allocate computing power resources to the target user according to the user allocation computing power resource information, and obtain user computing power resource information corresponding to the target user; Determine, according to the server identification information included in the user allocated computing power resource information, the value attribute information of the server corresponding to the server identification information; Determining the allocation time information included in the user computing power resource information as the target time information; Based on the target time information and the value attribute information, monitoring and processing the user computing power resource information includes: Based on the target time information, perform the following monitoring steps: According to the target time information, determining whether the current time information meets the preset time interval condition; In response to determining that the current time information satisfies the preset time interval condition, determining each idle time information of each computing power device resource included in the user computing power resource, wherein the idle time information in each idle time information is the duration of continuous idleness of the computing power device; Determine, based on the idle time information, whether there is computing power equipment resource information that meets a preset idle condition in the user computing power resource information; In response to determining that there is computing power device resource information that meets the preset idle condition in the user computing power resource information, determining each computing power device resource information that meets the preset idle condition in the user computing power resource information as each idle computing power device resource information, wherein each idle computing power device information represents each computing power device whose idle time is greater than the preset idle time information; Generate resource consumption score deduction ratio information according to the quantity of each computing power device resource information included in the user computing power resource information and the quantity of each idle computing power device resource information, wherein the resource consumption score deduction ratio information is the ratio of deducting the user resource consumption score of the target user; Deduction score information is generated according to the resource consumption score deduction ratio information and the value attribute information, wherein the deduction score information is the user resource consumption score information consumed by each GPU applied by the user within the preset time interval, wherein the deduction score corresponding to the value attribute information in the preset device value and deduction score relationship set is determined as the basic deduction score; the sum of the basic deduction score and the resource consumption score deduction ratio information is determined as the deduction multiplier information, and the deduction multiplier information represents the user resource consumption score information to be deducted per minute for a single GPU; the product of the deduction multiplier information and the number of each computing power device information included in the user computing power resource information is determined as the unit deduction score; the product of the unit deduction score and the preset value is determined as the deduction score information, the preset device value and deduction score relationship set represents the corresponding relationship between the value attribute information and the deduction score, the preset device value and deduction score relationship in the preset device value and deduction score relationship set includes the value attribute information and the deduction score corresponding to the value attribute information, and the unit deduction score is the user resource consumption score information consumed per minute for each GPU applied by the user; updating the user resource consumption score information of the target user according to the deduction score information; In response to determining that computing power device resource information that meets a preset warning condition exists in each computing power device resource information included in the user computing power resource information, computing power device idle warning information is sent to an associated terminal device; In response to determining that there is computing power device resource information that meets the preset shutdown condition in each computing power device resource information included in the user computing power resource information, the following steps are performed: Sending a computing device shutdown prompt message to the associated terminal device; Perform resource recovery processing on the computing power equipment resources corresponding to the idle computing power equipment resource information; Determine the idle computing power device resource information and the server identification information corresponding to the idle computing power device resource information as the remaining computing power resource information; Adding the remaining computing power resource information to the remaining computing power resource information set to update the remaining computing power resource information set; The current time information is used as the target time information, and based on the target time information, the monitoring step is performed again.

2. The method according to claim 1, wherein: The method also includes: In response to determining that the target user is a first-level administrator user and detecting that the target user performs an allocation operation on the user resource consumption score information, the following allocation steps are performed: In response to detecting the user identification information and the preset score information submitted by the target user, generating an assignment task corresponding to the user identification information and the preset score information; Adding the assigned task to an assigned task queue; In response to determining that the current time satisfies the preset allocation time condition, each allocation task in the obtained allocation task queue is executed.

3. The method according to claim 1, wherein: The method further comprises: In response to determining that the target user is a secondary administrator user and detecting that the target user performs an allocation operation on the user resource consumption score information, the following allocation steps are performed: updating the user resource consumption score information of the target user according to the user identification information and the pre-allocated score information submitted by the target user; The pre-allocated score information is determined as the user resource consumption score information corresponding to the user identification information.

4. The method according to claim 1, wherein: The computing power resource demand information includes server identification information and computing power device quantity information; and the determining the user allocation computing power resource information corresponding to the target user according to the computing power resource demand information, the user resource application authority information, the user resource consumption score information and the remaining computing power resource information set includes: Determine whether the computing power resource demand information meets the preset application conditions according to the user resource application authority information; In response to determining that the computing power resource demand information satisfies the preset application condition, determining whether the remaining computing power resource information set satisfies the preset allocation condition; In response to determining that the remaining computing power resource information set meets the preset allocation condition, the following steps are performed: determining the current time information as the allocation time information; According to the computing power resource demand information, the remaining computing power resource information set is updated; The allocation time information, the server identification information and the computing power device quantity information included in the computing power resource demand information are determined as the user allocation computing power resource information corresponding to the target user.

5. The method according to claim 4, wherein: The method further comprises: In response to determining that the computing power resource demand information does not meet the preset application condition, determining pre-allocated computing power resource demand information and computing power resource demand information to be preempted according to the user resource application authority information and the computing power resource demand information; determining the current time information as the allocation time information; According to the pre-allocated computing power resource demand information, the remaining computing power resource information set is updated; Determine the allocation time information, the server identification information and the computing power device quantity information included in the pre-allocated computing power resource demand information as the non-preemptive computing power resource information corresponding to the target user; Determine whether the remaining computing resource information set meets the preset computing resource preemption conditions according to the computing resource demand information to be preempted; In response to determining that the remaining computing power resource information set meets the preset preempting computing power resource condition, updating the remaining computing power resource information set according to the computing power resource demand information to be preempted; Determine the allocation time information, the server identification information and the computing power device quantity information included in the computing power resource demand information to be preempted as the preempted computing power resource information corresponding to the target user; Add the server identification information and computing power device quantity information included in the computing power resource demand information to be preempted as preempted device information to the preset preempted device information set; The non-preemptive computing power resource information and the preemptive computing power resource information are determined as user allocated computing power resource information corresponding to the target user.

6. The method according to claim 5, wherein: The method further comprises: In response to determining that the remaining computing power resource information does not satisfy the preset allocation condition, determining whether there is preset preemptive device information corresponding to the computing power resource demand information in the preset preemptive device information set; In response to determining that there is preset preemptive device information corresponding to the computing resource demand information in the preset preemptive device information set, the following steps are performed: Determine any one of the preset preemptive device information in each preset preemptive device information corresponding to the computing power resource demand information in the preset preemptive device information set as the pre-allocated device information; Backing up each task running in the server corresponding to the pre-allocated device information; Closing each task running in the server corresponding to the pre-allocated device information; Deleting the pre-allocated device information from the preset preempted device information set; determining the current time information as the allocation time information; The pre-allocated device information and the allocated time information are determined as user allocated computing power resource information corresponding to the target user.

7. A computing resource allocation device, comprising: A first display unit, configured to display a computing power device information interface and a navigation bar in response to detecting login information of a target user; A second display unit is configured to display a computing power resource details interface in response to detecting a selection operation of the target user on an entry control of the computing power resource details interface in the navigation bar; A third display unit is configured to display a computing power resource allocation interface in response to detecting a selection operation of the target user on an entry control of the computing power resource allocation interface in the navigation bar, wherein the computing power resource allocation interface includes a computing power resource application control; The allocation unit is configured to, in response to detecting the target user's selection operation on the computing power resource application control in the computing power resource allocation interface, perform the following allocation steps according to the computing power resource demand information submitted by the target user: obtain the user resource consumption score information, user resource application authority information and remaining computing power resource information set of the target user, wherein the remaining computing power resource information in the remaining computing power resource information set includes server identification information and each remaining computing power device information corresponding to the server identification information; determine the user allocation computing power resource information corresponding to the target user according to the computing power resource demand information, the user resource application authority information, the user resource consumption score information and the remaining computing power resource information set; allocate computing power resources to the target user according to the user allocation computing power resource information, and obtain the user computing power resource information corresponding to the target user; a value attribute information determining unit configured to determine the value attribute information of the server corresponding to the server identification information according to the server identification information included in the user allocated computing power resource information; a target time information determination unit, configured to determine the allocation time information included in the user computing power resource information as the target time information; The monitoring processing unit is configured to monitor and process the user computing power resource information based on the target time information and the value attribute information, including: Based on the target time information, perform the following monitoring steps: According to the target time information, determining whether the current time information meets the preset time interval condition; In response to determining that the current time information satisfies the preset time interval condition, determining each idle time information of each computing power device resource included in the user computing power resource, wherein the idle time information in each idle time information is the duration of continuous idleness of the computing power device; Determine, based on the idle time information, whether there is computing power equipment resource information that meets a preset idle condition in the user computing power resource information; In response to determining that there is computing power device resource information that meets the preset idle condition in the user computing power resource information, determining each computing power device resource information that meets the preset idle condition in the user computing power resource information as each idle computing power device resource information, wherein each idle computing power device information represents each computing power device whose idle time is greater than the preset idle time information; Generate resource consumption score deduction ratio information according to the quantity of each computing power device resource information included in the user computing power resource information and the quantity of each idle computing power device resource information, wherein the resource consumption score deduction ratio information is the ratio of deducting the user resource consumption score of the target user; Deduction score information is generated according to the resource consumption score deduction ratio information and the value attribute information, wherein the deduction score information is the user resource consumption score information consumed by each GPU applied by the user within the preset time interval, wherein the deduction score corresponding to the value attribute information in the preset device value and deduction score relationship set is determined as the basic deduction score; the sum of the basic deduction score and the resource consumption score deduction ratio information is determined as the deduction multiplier information, and the deduction multiplier information represents the user resource consumption score information to be deducted per minute for a single GPU; the product of the deduction multiplier information and the number of each computing power device information included in the user computing power resource information is determined as the unit deduction score; the product of the unit deduction score and the preset value is determined as the deduction score information, the preset device value and deduction score relationship set represents the corresponding relationship between the value attribute information and the deduction score, the preset device value and deduction score relationship in the preset device value and deduction score relationship set includes the value attribute information and the deduction score corresponding to the value attribute information, and the unit deduction score is the user resource consumption score information consumed per minute for each GPU applied by the user; updating the user resource consumption score information of the target user according to the deduction score information; In response to determining that computing power device resource information that meets a preset warning condition exists in each computing power device resource information included in the user computing power resource information, computing power device idle warning information is sent to an associated terminal device; In response to determining that there is computing power device resource information that meets the preset shutdown condition in each computing power device resource information included in the user computing power resource information, the following steps are performed: Sending a computing device shutdown prompt message to the associated terminal device; Perform resource recovery processing on the computing power equipment resources corresponding to the idle computing power equipment resource information; Determine the idle computing power device resource information and the server identification information corresponding to the idle computing power device resource information as the remaining computing power resource information; Adding the remaining computing power resource information to the remaining computing power resource information set to update the remaining computing power resource information set; The current time information is used as the target time information, and based on the target time information, the monitoring step is performed again.

8. An electronic device comprising: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images