Computing power scheduling method and device, network equipment and storage medium

Abstract

The invention discloses a computing power scheduling method and device, network equipment and a storage medium. The method comprises the following steps: acquiring resource information reported by at least one computing power node and request information of a target computing power service; determining a carbon use efficiency parameter corresponding to each computing power node according to the resource information; determining a scheduling strategy of the target computing power business based on the request information and the carbon use efficiency parameter; and scheduling and / or transferring the target computing power business by using the scheduling strategy.

Description

Technical Field

[0001] This application relates to the technical field of computing power networks, and particularly to a computing power scheduling method, device, network device, and storage medium. Background Art

[0002] The existing technology focuses on solving the selection of computing resources and network resources for computing power service scheduling, and there is little research on computing power service scheduling in combination with new energy, making it difficult to ensure the low-carbon operation of the computing power network. In response to this problem, there is currently no effective solution. Summary of the Invention

[0003] To solve the related technical problems, embodiments of this application provide a computing power scheduling method, device, network device, and storage medium.

[0004] The technical solution of the embodiments of this application is implemented as follows:

[0005] Embodiments of this application provide a computing power scheduling method, including:

[0006] Obtain the resource information reported by at least one computing power node and the request information of the target computing power service;

[0007] Determine the carbon usage efficiency parameter corresponding to each computing power node according to the resource information;

[0008] Determine the scheduling strategy of the target computing power service based on the request information and the carbon usage efficiency parameter;

[0009] Schedule and / or transfer the target computing power service using the scheduling strategy.

[0010] In the above solution, the resource information includes at least one of the following:

[0011] Computing power resource information;

[0012] Transmission resource information;

[0013] Energy resource information;

[0014] Energy efficiency resource information.

[0015] In the above solution, the energy resource information includes at least one of the following:

[0016] New energy type information;

[0017] New energy power generation time information;

[0018] New energy power generation amount information.

[0019] In the above solution, the energy efficiency resource information includes at least one of the following:

[0020] The energy efficiency level information of the computer room where the computing power node is located;

[0021] The power consumption information of the main equipment in the computer room where the computing power node is located.

[0022] In the above solution, when the resource information includes the energy efficiency level information and new energy power generation information of the computer room where the computing power node is located, the determining of the carbon usage efficiency parameter corresponding to each computing power node according to the resource information includes:

[0023] Determine the new energy usage ratio information according to the new energy power generation information;

[0024] Obtain the power carbon emission factor corresponding to the computing power node;

[0025] Determine the carbon usage efficiency parameter based on the new energy usage ratio information, the power carbon emission factor, and the energy efficiency level information.

[0026] In the above solution, the determining of the scheduling strategy for the target computing power service based on the request information and the carbon usage efficiency parameter includes:

[0027] Judge whether there is an unexecuted first computing power service request based on the request information;

[0028] When there is an unexecuted first computing power service request, obtain at least one first target computing power node that meets the resources required for the first computing power service request;

[0029] Determine the scheduling strategy for the target computing power service among the at least one first target computing power nodes according to the carbon usage efficiency parameter.

[0030] In the above solution, the determining of the scheduling strategy for the target computing power service among the at least one first target computing power nodes according to the carbon usage efficiency parameter includes:

[0031] Determine a second target computing power node with the optimal carbon usage efficiency among the at least one first target computing power nodes according to the carbon usage efficiency parameter;

[0032] Determine that the scheduling strategy is to schedule the first computing power service corresponding to the first computing power service request to the second target computing power node.

[0033] In the above solution, the method further includes:

[0034] When there is no unexecuted first computing power service request, obtain a third target computing power node that executes the first computing power service request and at least one fourth target computing power node that meets the resources required for the first computing power service request;

[0035] Determine whether there is a fourth target computing power node with a carbon usage efficiency higher than that of the third target computing power node;

[0036] In the case where there is a fourth target computing power node with a carbon usage efficiency higher than that of the third target computing power node among the at least one fourth target computing power node, obtain a fifth target computing power node with a carbon usage efficiency higher than that of the third target computing power node;

[0037] Determine that the scheduling policy is to transfer the computing power service executing the first computing power service request to the fifth target computing power node.

[0038] In the above solution, the method further includes:

[0039] Allocate computing power resource information to the second target computing power node or the fifth target computing power node.

[0040] In the above solution, the computing power resource information includes at least one of the following:

[0041] Computing power service information;

[0042] Occupied computing power resource information.

[0043] An embodiment of the present application also provides a computing power scheduling device, including:

[0044] An acquisition unit, configured to acquire resource information reported by at least one computing power node and request information of a target computing power service;

[0045] A first determination unit, configured to determine a carbon usage efficiency parameter corresponding to each computing power node according to the resource information;

[0046] A second determination unit, configured to determine a scheduling policy for the target computing power service based on the request information and the carbon usage efficiency parameter;

[0047] A scheduling unit, configured to schedule and / or transfer the target computing power service by using the scheduling policy.

[0048] An embodiment of the present application also provides a network device, including: a processor and a memory for storing a computer program that can run on the processor,

[0049] wherein, when the processor is used to run the computer program, it executes any step of the above method.

[0050] An embodiment of the present application also provides a computer-readable storage medium, storing executable instructions, which are used to implement any step of the above method when executed by a processor.

[0051] The computing power scheduling method, device, network device, and storage medium provided by the embodiments of the present application. Among them, the method includes: obtaining resource information reported by at least one computing power node and request information of a target computing power service; determining a carbon usage efficiency parameter corresponding to each computing power node according to the resource information; determining a scheduling strategy for the target computing power service based on the request information and the carbon usage efficiency parameter; using the scheduling strategy to schedule and / or transfer the target computing power service. The solution of the embodiments of the present application determines the carbon usage efficiency parameter corresponding to each computing power node through the resource information reported by at least one computing power node; determines the scheduling strategy for the target computing power service based on the request information of the target computing power service and the carbon usage efficiency parameter; uses the scheduling strategy to schedule and / or transfer the target computing power service; comprehensively considers factors such as energy usage and carbon usage efficiency in computing power scheduling, and selects computing power nodes based on the principle of optimal carbon usage efficiency, which can achieve the overall low-carbon operation of the computing power network. Description of the Drawings

[0052] Figure 1 It is a schematic diagram of the process of a computing power scheduling method provided by an embodiment of the present application;

[0053] Figure 2 It is a schematic diagram of the computing power network management platform initiating a request for a computing power node to report;

[0054] Figure 3 It is a schematic diagram of a computing power node actively reporting to the computing power network management platform;

[0055] Figure 4 It is a schematic diagram of the application process of a computing power scheduling method provided by an embodiment of the present application;

[0056] Figure 5 It is a schematic diagram of a computing power scheduling device provided by an embodiment of the present application;

[0057] Figure 6 It is a schematic diagram of a hardware entity structure of a network device in an embodiment of the present application. Detailed Embodiment

[0058] The present application will be further described in detail below with reference to the drawings and embodiments.

[0059] The computing power network needs to distribute the computing at one point to a line or a plane for processing, adding a lot of workloads such as task decomposition, data distribution, and data aggregation, which will inevitably increase the network resource occupation and consume a large amount of energy. First, for task decomposition, additional computing and storage resources are required. Secondly, the decomposed data is transmitted to each distributed computing center, which requires a large amount of network resources. Finally, the collection and processing of data also require computing, storage, and network resources. These new workloads will bring a large amount of energy consumption, posing a severe challenge to the green and low-carbon development of the computing power network.

[0060] New energy is regarded as an important solution to energy and environmental problems and to achieve the "dual carbon" goal. There are mainly two ways to apply new energy to the computing power network. In areas with suitable climate conditions, building small-scale wind power, rooftop photovoltaic and other power generation devices is encouraged; in areas where green electricity prices are advantageous, purchasing low-cost green electricity through power trading is encouraged. With the development of a new power system with new energy as the main body, new energy will be more widely applied to the computing power network.

[0061] Related technologies focus on solving the selection of computing resources and network resources for computing power service scheduling, and there is little research on combining new energy for computing power service scheduling, making it difficult to ensure the low-carbon operation of the computing power network. Therefore, there is an urgent need for a computing power network scheduling method that takes into account new energy and energy efficiency information, which can not only ensure the low-carbon operation of the computing power network, but also improve the utilization efficiency of new energy and promote the large-scale consumption of new energy.

[0062] On the one hand, related technologies focus on solving the selection of computing resources and network resources during the computing power service scheduling process, and there is little research on the computing power network scheduling that combines new energy. On the other hand, in terms of computing power service migration, existing technologies usually consider the computing migration that occurs when users change computing power nodes when they are in a mobile state, without considering the computing migration based on the new energy utilization efficiency of the computing power nodes.

[0063] Based on this, comprehensively considering the factors of new energy use and energy efficiency in the computing power network scheduling and selecting computing power nodes based on the principle of optimal carbon use efficiency can achieve the overall low-carbon operation of the computing power network.

[0064] The embodiment of this application provides a computing power scheduling method, which is applied to a network device. The functions implemented by this method can be realized by a processor in the network device calling program code. Of course, the program code can be stored in a computer storage medium. It can be seen that this network device includes at least a processor and a storage medium. As an example, this network device can be a computing network management platform, a gateway, a server, a computer, etc.

[0065] Figure 1 FIG. [ID] is a schematic diagram of the process of a computing power scheduling method provided by an embodiment of this application; as Figure 1 shown, this method includes:

[0066] Step 101: Obtain the resource information reported by at least one computing power node and the request information of the target computing power service;

[0067] Step 102: Determine the carbon use efficiency parameter corresponding to each computing power node according to the resource information;

[0068] Step 103: Determine the scheduling strategy of the target computing power service based on the request information and the carbon use efficiency parameter;

[0069] Step 104: Schedule and / or transfer the target computing power service by using the scheduling policy.

[0070] In this embodiment, the specific scheduling content of the computing power scheduling method can be determined according to the actual situation and is not limited herein. As an example, the computing power scheduling method can specifically be a computing power network scheduling method.

[0071] In step 101, the specific number of the at least one computing power node can be determined according to the actual situation and is not limited herein. The resource information can be determined according to the actual situation and is not limited herein. As an example, the resource information can include at least one of the following: computing power resource information; transmission resource information; energy resource information; energy efficiency resource information. In practical applications, the resource information can also be referred to as resource status information; the computing power resource information can be abbreviated as computing power resources; the transmission resource information can be abbreviated as transmission resources; the energy resource information can be abbreviated as energy resources; the energy efficiency resource information can be abbreviated as energy efficiency information.

[0072] Obtaining the resource information reported by at least one computing power node can be through periodic reporting actively initiated by at least one computing power node, or event-based reporting, such as a change in energy resources, etc., or can be initiated by the computing power network management platform to request the computing power node to report.

[0073] The request information of the target computing power service needs to be determined according to the actual requirements and is not limited herein. As an example, the request information of the target computing power service can be the computing power service requirements of the computing power resources and transmission resources of the computing power node.

[0074] In step 102, the specific determination process of determining the carbon usage efficiency parameter corresponding to each computing power node according to the resource information can be determined according to the actual situation and is not limited herein; as an example, the resource information can include the energy efficiency level information and new energy power generation information of the computer room where the computing power node is located. The determining the carbon usage efficiency parameter corresponding to each computing power node according to the resource information can include: determining the new energy usage ratio information according to the new energy power generation information; obtaining the power carbon emission factor corresponding to the computing power node; determining the carbon usage efficiency parameter based on the new energy usage ratio information, the power carbon emission factor, and the energy efficiency level information. In practical applications, the carbon usage efficiency parameter can be abbreviated as carbon usage efficiency (CUE).

[0075] In step 103, the specific determination process in determining the scheduling strategy of the target computing power service based on the request information and the carbon usage efficiency parameter can be determined according to the actual situation and is not limited herein; as an example, the determination of the scheduling strategy of the target computing power service based on the request information and the carbon usage efficiency parameter may include determining whether there is an unexecuted first computing power service request based on the request information; in the case where there is an unexecuted first computing power service request, obtaining at least one first target computing power node that meets the resources required for the first computing power service request; and determining the scheduling strategy of the target computing power service among the at least one first target computing power node according to the carbon usage efficiency parameter.

[0076] In practical applications, the scheduling strategy can also be referred to as a scheduling decision. The computing power and network management platform makes a scheduling decision based on computing power service requests, new energy and energy efficiency information of each computing power node, and forms computing power resource allocation information based on the principle of optimal carbon usage efficiency.

[0077] In step 104, the scheduling and / or transfer of the target computing power service using the scheduling strategy can be understood as scheduling the target computing power service using the scheduling strategy; or, transferring the target computing power service using the scheduling strategy; or, scheduling and transferring the target computing power service using the scheduling strategy.

[0078] In practical applications, scheduling and / or transferring the target computing power service using the scheduling strategy can be achieved by the computing power and network management platform sending the computing power resource allocation information to the target computing power node, where the computing power resource allocation information includes computing power service information, occupied computing power resources, etc. The computing power and network management platform sends the computing power resource allocation information to the target computing power node, establishes a connection between the computing power service and the target computing power node, and executes the computing power service. If it is a computing power service transfer, the computing power and network management platform also needs to notify the source computing power node to send the necessary computing power service data to the target computing power node and end the computing power service.

[0079] In one embodiment, the resource information includes at least one of the following:

[0080] Computing power resource information;

[0081] Transmission resource information;

[0082] Energy resource information;

[0083] Energy efficiency resource information.

[0084] Among them, the computing power resource information, the transmission resource information, the energy resource information, and the energy efficiency resource information can all be determined according to actual situations and are not limited herein. As an example, the energy resource information may include at least one of the following: new energy type information; new energy power generation time information; new energy power generation amount information, etc. The energy efficiency resource information may include at least one of the following: the energy efficiency level information of the computer room where the computing power node is located; the main equipment power consumption information of the computer room where the computing power node is located, etc. The computing power resource information may include at least one of the following: idle computing power resources, computing power resource utilization rate, etc.; the transmission resource information may include transmission bandwidth, etc.

[0085] The computing power resource information can be abbreviated as computing power resources; the transmission resource information can be abbreviated as transmission resources; the energy resource information can be abbreviated as energy resources; the energy efficiency resource information can be abbreviated as energy efficiency information. The energy resource information can be abbreviated as energy resources; the new energy type information can be abbreviated as new energy types; the new energy power generation time information can be abbreviated as new energy power generation time; the new energy power generation amount information can be abbreviated as power generation amount. The energy efficiency level information of the computer room where the computing power node is located can be abbreviated as the PUE of the computer room where it is located; the main equipment power consumption information of the computer room where the computing power node is located can be abbreviated as main equipment power consumption.

[0086] In practical applications, energy resources include new energy power generation time, power generation amount, new energy types, etc.; energy efficiency information includes the PUE of the computer room where it is located, main equipment power consumption, etc.; computing power resources include idle computing power resources, computing power resource utilization rate, etc.; transmission resources include transmission bandwidth, etc.

[0087] In one embodiment, the energy resource information includes at least one of the following:

[0088] New energy type information;

[0089] New energy power generation time information;

[0090] New energy power generation amount information.

[0091] In this embodiment, the energy resource information can be abbreviated as energy resources; the new energy type information can be abbreviated as new energy types; the new energy power generation time information can be abbreviated as new energy power generation time; the new energy power generation amount information can be abbreviated as power generation amount.

[0092] In practical applications, the energy resources may include new energy power generation time, power generation amount, new energy types, etc.

[0093] In one embodiment, the energy efficiency resource information includes at least one of the following:

[0094] The energy efficiency level information of the computer room where the computing power node is located;

[0095] The power consumption information of the main equipment in the computer room where the computing power node is located.

[0096] In this embodiment, the energy efficiency resource information can be abbreviated as energy efficiency information; the energy efficiency level information of the computer room where the computing power node is located can be abbreviated as the PUE of the computer room where it is located; the power consumption information of the main equipment in the computer room where the computing power node is located can be abbreviated as the power consumption of the main equipment.

[0097] In practical applications, the energy efficiency information includes the PUE of the computer room where it is located, the power consumption of the main equipment, etc.

[0098] In one embodiment, when the resource information includes the energy efficiency level information and new energy power generation information of the computer room where the computing power node is located, determining the carbon usage efficiency parameter corresponding to each computing power node according to the resource information includes:

[0099] Determine the new energy usage ratio information according to the new energy power generation information;

[0100] Obtain the power carbon emission factor corresponding to the computing power node;

[0101] Determine the carbon usage efficiency parameter based on the new energy usage ratio information, the power carbon emission factor, and the energy efficiency level information.

[0102] In this embodiment, the specific determination process of determining the new energy usage ratio information according to the new energy power generation information can be determined according to the actual situation and is not limited here. As an example, determining the new energy usage ratio information according to the new energy power generation information can be to obtain the new energy power generation time based on the new energy power generation information, and then determine the new energy usage ratio information based on the new energy power generation time.

[0103] The energy efficiency level information of the computer room where the computing power node is located can be denoted as PUE, which can represent the energy efficiency level of the computer room where the computing power node is located, and is equal to the total power consumption of the computer room / the power consumption of IT equipment. Generally speaking, PUE is a value greater than 1. The lower the PUE value, the lower the energy consumption other than for IT equipment in the data center. It is sourced from the information reported by the computing power node.

[0104] The new energy usage ratio information can be denoted as RER, which can represent the new energy usage ratio and can be calculated based on the new energy power generation time reported by the computing power node.

[0105] The power carbon emission factor can be denoted as α, representing the power carbon emission factor. For commercial power, α can refer to the average CO2 emission factor of the Chinese regional power grid (kg CO2 / kWh), and for new energy represented by distributed photovoltaics, α = 0.

[0106] The carbon usage efficiency parameter can be abbreviated as CUE.

[0107] In this embodiment, it is mainly considered that the factors affecting the carbon usage efficiency include the energy efficiency level of the computer room, the proportion of new energy used, the energy carbon emission factor, etc. Generally, the carbon usage of computing nodes using new energy is lower. Specifically, the carbon usage efficiency is defined as CUE (Carbon Usage Effectiveness, CUE), and the calculation process can refer to the following formula (1):

[0108] CUE = PUE * α * (1 - RER) (1)

[0109] Wherein,

[0110] PUE represents the energy efficiency level of the computer room where the computing node is located, which is equal to the total power consumption of the computer room / the power consumption of IT equipment. Generally speaking, PUE is a value greater than 1. The lower the PUE value, the lower the energy consumption outside the IT equipment in the data center. It is derived from the information reported by the computing node.

[0111] α represents the power carbon emission factor. For commercial power, α can refer to the average CO2 emission factor of the Chinese regional power grid (kgCO2 / kWh). For new energy represented by distributed photovoltaics, α = 0.

[0112] RER represents the proportion of new energy used, which is calculated based on the new energy power generation time reported by the computing node.

[0113] In formula (1), CUE is positively correlated with PUE and α, and negatively correlated with RER. Exemplarily, the smaller the PUE, the higher the energy efficiency of the computer room, and the smaller the CUE; the smaller the α, the less the carbon emission per unit of electricity, and the smaller the CUE; the higher the RER, the higher the proportion of new energy used, and the smaller the CUE.

[0114] This method can ensure that the new energy power generation meets the power consumption of the computing service, or if it is insufficient for the power consumption of the computing service, it can use the new energy to the greatest extent, and it is more applicable to the scenario of hybrid power supply of new energy and commercial power.

[0115] In one embodiment, determining the scheduling strategy of the target computing service based on the request information and the carbon usage efficiency parameter includes:

[0116] Judging whether there is an unexecuted first computing service request based on the request information;

[0117] In the case of an unexecuted first computing service request, obtaining at least one first target computing node that meets the resources required by the first computing service request;

[0118] Determine the scheduling strategy of the target computing power service among the at least one first target computing power nodes according to the carbon usage efficiency parameter.

[0119] In the embodiment of the present application, judging whether there is an unexecuted first computing power service request based on the request information may be that the computing network management platform judges whether there is an unexecuted computing power service request based on the request information.

[0120] In the case where there is an unexecuted first computing power service request, obtain at least one first target computing power node that meets the resources required for the first computing power service request; among them, obtaining at least one first target computing power node that meets the resources required for the first computing power service request can be understood as selecting a computing power node powered by new energy to meet the electricity consumption of the computing power service, that is, the power generation amount is greater than or equal to the service power consumption and the power generation time is greater than or equal to the service duration. As an example, it can be understood that there is an unexecuted first computing power service request, and a computing power node powered by new energy to meet the electricity consumption of the computing power service is selected, that is, the power generation amount is greater than or equal to the service power consumption and the power generation time is greater than or equal to the service duration.

[0121] The specific determination process of determining the scheduling strategy of the target computing power service among the at least one first target computing power nodes according to the carbon usage efficiency parameter can be determined according to the actual situation and will not be limited here. As an example, determining the scheduling strategy of the target computing power service among the at least one first target computing power nodes according to the carbon usage efficiency parameter may include determining a second target computing power node with the optimal carbon usage efficiency among the at least one first target computing power nodes according to the carbon usage efficiency parameter; determining the scheduling strategy as scheduling the first computing power service corresponding to the first computing power service request to the second target computing power node.

[0122] In practical applications, the computing network management platform makes scheduling decisions based on the computing power service request, the new energy and energy efficiency information of each computing power node, and forms computing power resource allocation information based on the principle of optimal carbon usage efficiency. Specifically, in the first step, it is judged whether there is an unexecuted computing power service request. If so, enter the resource allocation process for new computing power services; otherwise, enter the resource allocation process for running computing power services. In the second step, selection of computing power nodes for new computing power services. First, the computing power resources and transmission resources of the computing power nodes both meet the computing power service requirements; if all computing power nodes do not meet the requirements, wait for the next cycle or consider decomposing the computing task, which will not be elaborated here. Secondly, select a computing power node powered by new energy to meet the electricity consumption of the computing power service, that is, the power generation amount is greater than or equal to the service power consumption and the power generation time is greater than or equal to the service duration; if all do not meet the requirements, enter the next step. Thirdly, select the computing power node with the minimum carbon usage efficiency, or to save the amount of calculation, the computing power node with a carbon usage efficiency less than the preset threshold can be preferentially selected.

[0123] Based on this, in one embodiment, determining the scheduling policy of the target computing power service among the at least one first target computing power nodes according to the carbon usage efficiency parameter includes:

[0124] Determining a second target computing power node with the optimal carbon usage efficiency among the at least one first target computing power nodes according to the carbon usage efficiency parameter;

[0125] Determining the scheduling policy as scheduling the first computing power service corresponding to the first computing power service request to the second target computing power node.

[0126] The computing power resources and transmission resources of the computing power node both meet the computing power service requirements; if all computing power nodes do not meet the requirements, wait for the next cycle or consider decomposing the computing task, which will not be elaborated here.

[0127] In one embodiment, the method further includes:

[0128] In the case where there is no unexecuted first computing power service request, obtaining a third target computing power node for executing the first computing power service request and at least one fourth target computing power node that meets the resources required for the first computing power service request;

[0129] Judging whether there is a carbon usage efficiency higher than that of the third target computing power node among the at least one fourth target computing power nodes;

[0130] In the case where there is a carbon usage efficiency higher than that of the third target computing power node among the at least one fourth target computing power nodes, obtaining a fifth target computing power node with a carbon usage efficiency higher than that of the third target computing power node;

[0131] Determining the scheduling policy as transferring the computing power service for executing the first computing power service request to the fifth target computing power node.

[0132] In this embodiment, in the case where there is no unexecuted first computing power service request, it can be understood as entering the resource allocation process of the running computing power service, that is, the transfer of the computing power node of the running computing power service.

[0133] In practical applications, the specific process of the transfer of computing power nodes that have already run computing power services is as follows: First, the computing power resources and transmission resources of other computing power nodes (other than the computing power nodes where the computing power services are running) meet the computing power service requirements; otherwise, the process ends. Second, select other computing power nodes whose carbon usage efficiency decreases before and after the transfer of the computing power service, and the change amount is higher than the threshold. Herein, the threshold can be set by considering the carbon usage efficiency reduction benefit and transfer cost of the computing power service transfer. Third, compare the carbon usage efficiency CUEworking_node of the computing power node where the computing power service is running with the carbon usage efficiency CUEother_node of other computing power nodes. If the following conditions are met, and the following conditions can refer to Equation (2) and Equation (3), then transfer the running computing power service from the current computing power node to the computing power node with better carbon usage efficiency.

[0134] CUEworking_node > CUEother_node (2)

[0135] |CUEworking_node - CUEother_node| > CUEthreshold (3)

[0136] This embodiment calculates and migrates based on the carbon usage efficiency of the computing power node, makes full use of the computing power nodes with a high proportion of new energy, and the original computing power node can be put into sleep to reduce energy consumption, which is more conducive to the overall low-carbon operation of the computing power network.

[0137] In one embodiment, the method further includes:

[0138] Allocating computing power resource information to the second target computing power node or the fifth target computing power node.

[0139] In this embodiment, allocating computing power resource information to the second target computing power node or the fifth target computing power node can be understood as the computing network management platform sending the computing power resource information to the second target computing power node or the fifth target computing power node.

[0140] In practical applications, the computing network management platform sends the computing power resource allocation information to the target computing power node.

[0141] In one embodiment, the computing power resource information includes at least one of the following:

[0142] Computing power service information;

[0143] Occupied computing power resource information.

[0144] In this embodiment, the computing network management platform sends the computing power resource allocation information to the target computing power node. The computing power resource allocation information includes computing power service information, occupied computing power resources, etc. The computing network management platform sends the computing power resource allocation information to the target computing power node, establishes the connection between the computing power service and the target computing power node, and executes the computing power service. If it is a transfer of the computing power service, the computing network management platform also needs to notify the source computing power node to send the necessary computing power service data to the target computing power node and end the computing power service.

[0145] For the convenience of understanding, an example is given here. The computing power scheduling method is specifically the computing network scheduling method, and the network device is specifically the computing network management platform. The specific process is as follows:

[0146] 1. The computing network management platform receives the resource status information reported by the computing power node. The resource status information includes but is not limited to:

[0147] Energy resources, including new energy power generation time, power generation amount, new energy type, etc.;

[0148] Energy efficiency information, including PUE of the computer room where it is located, power consumption of the main equipment, etc.;

[0149] Computing power resources, including idle computing power resources, computing power resource utilization rate, etc.; Transmission resources, including transmission bandwidth, etc.

[0150] The resource status information can be actively reported periodically by the computing power node, or reported eventually, such as when the energy resources change, or can be requested by the computing network management platform to require the computing power node to report. This content can be combined with Figure 2 and Figure 3 for understanding. Figure 2 is a schematic diagram of the computing network management platform requesting the computing power node to report. In Figure 2 , the Computing Network Management Platform requests the RESOURCE_STATUS_REQUEST from the Computing power Node, and the Computing power Node returns the RESOURCE_STATUS_RESPONSE to the Computing Network Management Platform. Figure 3 is a schematic diagram of the computing power node actively reporting to the computing network management platform. In Figure 3 , the Computing power Node conducts RESOURCE_STATUS_REPORT to the Computing Network Management Platform.

[0151] 2. The computing and network management platform makes scheduling decisions based on computing power service requests, new energy and energy efficiency information of each computing power node, and forms computing power resource allocation information based on the principle of optimal carbon usage efficiency.

[0152] 2.1 Determine whether there are unexecuted computing power service requests. If there are, enter the new computing power service resource allocation process; otherwise, enter the resource allocation process for running computing power services.

[0153] 2.2 Selection of computing power nodes for new computing power services.

[0154] First, the computing power resources and transmission resources of the computing power nodes both meet the computing power service requirements; if all computing power nodes do not meet the requirements, wait for the next cycle or consider decomposing the computing tasks, which will not be elaborated here.

[0155] Second, select computing power nodes powered by new energy that meet the power consumption of the computing power service, that is, the power generation is greater than or equal to the service power consumption and the power generation time is greater than or equal to the service duration; if all do not meet the requirements, proceed to the next step.

[0156] Third, select the computing power node with the minimum carbon usage efficiency, or to save the amount of calculation, you can preferentially select the computing power node with a carbon usage efficiency less than the preset threshold.

[0157] The factors affecting carbon usage efficiency include factors such as the energy efficiency level of the computer room, the proportion of new energy use, and the energy carbon emission factor. Generally, the carbon usage of computing power nodes using new energy is lower. Specifically, for carbon usage efficiency (Carbon Usage Effectiveness, CUE), the calculation process can refer to formula (1) above.

[0158] Among them,

[0159] PUE represents the energy efficiency level of the computer room where the computing power node is located, equal to the total power consumption of the computer room / the power consumption of IT equipment. Generally speaking, PUE is a value greater than 1. The lower the PUE value, the lower the energy consumption outside the IT equipment in the data center. It comes from the information reported by the computing power node.

[0160] α represents the power carbon emission factor. For commercial power, α can refer to the average CO2 emission factor of the Chinese regional power grid (kgCO2 / kWh). For new energy represented by distributed photovoltaic, α = 0.

[0161] RER represents the proportion of new energy usage, which is calculated based on the new energy generation time reported by the computing power nodes. In Equation (1), CUE is positively correlated with PUE and α, and negatively correlated with RER. Exemplarily, the smaller the PUE, the higher the energy efficiency of the computer room and the smaller the CUE; the smaller the α, the less carbon emissions per unit of electricity and the smaller the CUE; the higher the RER, the higher the proportion of new energy usage and the smaller the CUE.

[0162] This method can ensure that the new energy generation meets the electricity demand of the computing power service, or if it is insufficient for the computing power service, it can use new energy to the greatest extent possible. It is more applicable to the scenario of hybrid power supply of new energy and commercial power.

[0163] 2.3 Transfer of computing power nodes with running computing power services.

[0164] First, the computing power resources and transmission resources of other computing power nodes (other than the computing power node where the computing power service is running) meet the computing power service requirements; otherwise, end.

[0165] Secondly, select other computing power nodes whose carbon usage efficiency decreases after the transfer of the computing power service and the change amount is higher than the threshold. Among them, the threshold can be set considering the reduction benefit of carbon usage efficiency and the transfer cost when the computing power service is transferred.

[0166] Compare the carbon usage efficiency CUEworking_node of the computing power node where the computing power service is running with the carbon usage efficiency CUEother_node of other computing power nodes. If the following conditions are met, referring to Equations (2) and (3) above, the running computing power service can be transferred from the current computing power node to the computing power node with better carbon usage efficiency.

[0167] This method calculates and migrates based on the carbon usage efficiency of the computing power nodes, makes full use of the computing power nodes with a high proportion of new energy, and the original computing power node can be put into sleep to reduce energy consumption, which is more conducive to the overall low-carbon operation of the computing power network.

[0168] The computing power network management platform sends the computing power resource allocation information to the target computing power node. Among them, the computing power resource allocation information includes computing power service information, occupied computing power resources, etc.

[0169] The computing power network management platform sends the computing power resource allocation information to the target computing power node, establishes the connection between the computing power service and the target computing power node, and executes the computing power service. If it is a transfer of the computing power service, the computing power network management platform also needs to notify the source computing power node to send the necessary computing power service data to the target computing power node and end the computing power service.

[0170] For better understanding, the specific process of this embodiment can be combined with Figure 4 for understanding, Figure 4 is a schematic application process diagram of a computing power scheduling method provided by an embodiment of this application;

[0171] In the embodiments of the present application, by obtaining the new energy and energy efficiency information of computing power nodes, etc., based on the principle of optimal carbon usage efficiency, the computing power services are scheduled and / or the computing power services are transferred to the computing power nodes with a high proportion of new energy, improving the utilization efficiency of new energy and realizing the low-carbon operation of the overall computing power network. On the one hand, it can comprehensively consider the size of new energy generation and the size of energy efficiency of computing power nodes, and is applicable to the scenario of hybrid power supply of new energy and commercial power. On the other hand, it obtains the energy resources, energy efficiency information, etc. of the computing power nodes. The energy resources are mainly the new energy generation time and power generation amount, etc., and the energy efficiency information is mainly the PUE of the computer room, etc.

[0172] Compared with the prior art, in the present application, by obtaining the new energy and energy efficiency information of computing power nodes, etc., based on the principle of optimal carbon usage efficiency, the computing power services are scheduled and / or the computing power services are transferred to the computing power nodes with a high proportion of new energy, improving the utilization efficiency of new energy and realizing the low-carbon operation of the overall computing power network.

[0173] To implement the method of the embodiments of the present application, the embodiments of the present application also provide a computing power scheduling device 500, which is set on a network device, such as Figure 5 as shown Figure 5 is a schematic diagram of a computing power scheduling device provided by an embodiment of the present application, including:

[0174] An obtaining unit 501, configured to obtain resource information reported by at least one computing power node and request information of a target computing power service;

[0175] A first determining unit 502, configured to determine a carbon usage efficiency parameter corresponding to each computing power node according to the resource information;

[0176] A second determining unit 503, configured to determine a scheduling strategy for the target computing power service based on the request information and the carbon usage efficiency parameter;

[0177] A scheduling unit 504, configured to schedule and / or transfer the target computing power service by using the scheduling strategy.

[0178] Here, in one embodiment, the resource information includes at least one of the following:

[0179] Computing power resource information;

[0180] Transmission resource information;

[0181] Energy resource information;

[0182] Energy efficiency resource information.

[0183] Here, in one embodiment, the energy resource information includes at least one of the following:

[0184] New energy type information;

[0185] New energy power generation time information;

[0186] New energy power generation amount information.

[0187] Here, in one embodiment, the energy efficiency resource information includes at least one of the following:

[0188] The energy efficiency level information of the computer room where the computing power node is located;

[0189] The main equipment power consumption information of the computer room where the computing power node is located.

[0190] Here, in one embodiment, when the resource information includes the energy efficiency level information of the computer room where the computing power node is located and the new energy power generation amount information, the first determination unit 502 is further configured to determine new energy usage ratio information according to the new energy power generation amount information; obtain the power carbon emission factor corresponding to the computing power node; and determine the carbon usage efficiency parameter based on the new energy usage ratio information, the power carbon emission factor, and the energy efficiency level information.

[0191] Here, in one embodiment, the second determination unit 503 is further configured to determine whether there is an unexecuted first computing power service request based on the request information; when there is an unexecuted first computing power service request, obtain at least one first target computing power node that meets the resources required for the first computing power service request; and determine the scheduling strategy of the target computing power service among the at least one first target computing power nodes according to the carbon usage efficiency parameter.

[0192] Here, in one embodiment, the second determination unit 503 is further configured to determine a second target computing power node with the optimal carbon usage efficiency among the at least one first target computing power nodes according to the carbon usage efficiency parameter; and determine the scheduling strategy as scheduling the first computing power service corresponding to the first computing power service request to the second target computing power node.

[0193] Here, in one embodiment, the apparatus 500 further includes a judgment unit; wherein,

[0194] The obtaining unit 501 is further configured to obtain a third target computing power node for executing the first computing power service request and at least one fourth target computing power node that meets the resources required for the first computing power service request when there is no unexecuted first computing power service request;

[0195] The judgment unit is configured to judge whether there is a carbon usage efficiency higher than that of the third target computing power node among the at least one fourth target computing power nodes;

[0196] The obtaining unit 501 is further configured to, when there is a fifth target computing power node with a carbon usage efficiency higher than that of the third target computing power node among the at least one fourth target computing power node, obtain the fifth target computing power node with a carbon usage efficiency higher than that of the third target computing power node.

[0197] The scheduling unit 504 is further configured to determine that the scheduling policy is to transfer the computing power service for executing the first computing power service request to the fifth target computing power node.

[0198] In an embodiment, the apparatus 500 further includes an allocation unit configured to allocate computing power resource information to the second target computing power node or the fifth target computing power node.

[0199] In an embodiment, the computing power resource information includes at least one of the following:

[0200] Computing power service information;

[0201] Occupied computing power resource information.

[0202] It should be noted that: when the computing power scheduling apparatus provided in the above embodiment performs computing power scheduling, only the division of the above program modules is used for illustration. In practical applications, the above processing can be allocated to different program modules according to needs, that is, the internal structure of the apparatus is divided into different program modules to complete all or part of the above-described processing. In addition, the computing power scheduling apparatus provided in the above embodiment and the embodiment of the computing power scheduling method belong to the same concept. For the specific implementation process, please refer to the method embodiment, which will not be elaborated here.

[0203] Based on the hardware implementation of the above program modules, an embodiment of the present application further provides a network device, including a memory and a processor. The memory stores a computer program that can run on the processor, and when the processor executes the program, it implements the steps in the computing power scheduling method provided in the above embodiment.

[0204] Correspondingly, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps in the computing power scheduling method provided in the above embodiment.

[0205] It should be pointed out here that: the descriptions of the above storage medium and device embodiments are similar to those of the above method embodiments and have beneficial effects similar to those of the method embodiments. For the technical details not disclosed in the storage medium and device embodiments of the present application, please refer to the description of the method embodiments of the present application for understanding.

[0206] It should be noted that Figure 6 is a schematic diagram of a hardware entity structure of the network device in the embodiment of the present application. As Figure 6As shown, the hardware entities of the network device 600 include: a processor 601 and a memory 603. Optionally, the network device 600 may further include a communication interface 602.

[0207] It can be understood that the memory 603 can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM, Read Only Memory), a programmable read-only memory (PROM, Programmable Read-Only Memory), an erasable programmable read-only memory (EPROM, Erasable Programmable Read-Only Memory), an electrically erasable programmable read-only memory (EEPROM, Electrically Erasable Programmable Read-Only Memory), a ferromagnetic random access memory (FRAM, ferromagnetic random access memory), a flash memory (Flash Memory), a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM, Compact Disc Read-Only Memory); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM, Random Access Memory), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as a static random access memory (SRAM, Static Random Access Memory), a synchronous static random access memory (SSRAM, Synchronous Static Random Access Memory), a dynamic random access memory (DRAM, Dynamic Random Access Memory), a synchronous dynamic random access memory (SDRAM, Synchronous Dynamic Random Access Memory), a double data rate synchronous dynamic random access memory (DDR SDRAM, Double Data Rate Synchronous Dynamic Random Access Memory), an enhanced synchronous dynamic random access memory (ESDRAM, Enhanced Synchronous Dynamic Random Access Memory), a sync link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), a direct rambus random access memory (DRRAM, Direct Rambus Random Access Memory).The memory 603 described in the embodiments of the present application is intended to include, but is not limited to, these and any other suitable types of memories.

[0208] The methods disclosed in the embodiments of the present application above can be applied to, or implemented by, the processor 601. The processor 601 may be an integrated circuit chip with signal processing capabilities. In the implementation process, the steps of the above methods can be completed by the integrated logic circuit in hardware or instructions in software form in the processor 601. The above-mentioned processor 601 may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 601 can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or any conventional processor, etc. Combining the steps of the methods disclosed in the embodiments of the present application can be directly embodied as being executed and completed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium, and this storage medium is located in the memory 603. The processor 601 reads the information in the memory 603 and combines its hardware to complete the steps of the foregoing methods.

[0209] In an exemplary embodiment, the device can be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontroller units (MCUs), microprocessors, or other electronic components for executing the foregoing methods.

[0210] It should be understood that the "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, the appearances of "in one embodiment" or "in an embodiment" throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics may be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the magnitudes of the sequence numbers of the above processes do not mean the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application. The sequence numbers of the embodiments of the present application above are only for description and do not represent the advantages or disadvantages of the embodiments.

[0211] It should be noted that in this text, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including such element.

[0212] The methods disclosed in several method embodiments provided by the present application can be arbitrarily combined without conflict to obtain new method embodiments.

[0213] The features disclosed in several product embodiments provided by the present application can be arbitrarily combined without conflict to obtain new product embodiments.

[0214] The features disclosed in several method or device embodiments provided by the present application can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0215] As mentioned above, the above is only the implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A computing power scheduling method, characterized in that, Including: Obtaining resource information reported by at least one computing power node and request information of a target computing power service; Determining a carbon usage efficiency parameter corresponding to each computing power node according to the resource information; Determining a scheduling strategy for the target computing power service based on the request information and the carbon usage efficiency parameter; Scheduling and / or transferring the target computing power service by using the scheduling strategy.

2. The method according to claim 1, wherein The resource information includes at least one of the following: Computing power resource information; Transmission resource information; Energy resource information; Energy efficiency resource information.

3. The method according to claim 2, characterized in that, The energy resource information includes at least one of the following: New energy type information; New energy power generation time information; New energy power generation amount information.

4. The method according to claim 2, wherein The energy efficiency resource information includes at least one of the following: Energy efficiency level information of the computer room where the computing power node is located; Main equipment power consumption information of the computer room where the computing power node is located.

5. The method according to claim 1, characterized in that, When the resource information includes the energy efficiency level information of the computer room where the computing power node is located and the new energy power generation amount information, the determining a carbon usage efficiency parameter corresponding to each computing power node according to the resource information includes: Determining new energy usage ratio information according to the new energy power generation amount information; Obtaining the power carbon emission factor corresponding to the computing power node; Determining the carbon usage efficiency parameter based on the new energy usage ratio information, the power carbon emission factor and the energy efficiency level information.

6. The method according to claim 4, wherein The determining a scheduling strategy for the target computing power service based on the request information and the carbon usage efficiency parameter includes: Judging whether there is an unexecuted first computing power service request based on the request information; When there is an unexecuted first computing power service request, obtaining at least one first target computing power node required for the first computing power service request; Determining the scheduling strategy for the target computing power service among the at least one first target computing power nodes according to the carbon usage efficiency parameter.

7. The method according to claim 6, characterized in that, The determining the scheduling strategy for the target computing power service among the at least one first target computing power nodes according to the carbon usage efficiency parameter includes: Determining a second target computing power node with the optimal carbon usage efficiency among the at least one first target computing power nodes according to the carbon usage efficiency parameter; Determining the scheduling strategy as scheduling the first computing power service corresponding to the first computing power service request to the second target computing power node.

8. The method according to claim 7, characterized in that, The method further includes: When there is no unexecuted first computing power service request, obtaining a third target computing power node executing the first computing power service request and at least one fourth target computing power node required for the first computing power service request; Judging whether there is a computing power node with a carbon usage efficiency higher than that of the third target computing power node among the at least one fourth target computing power nodes; When there is a computing power node with a carbon usage efficiency higher than that of the third target computing power node among the at least one fourth target computing power nodes, obtaining a fifth target computing power node with a carbon usage efficiency higher than that of the third target computing power node; Determining the scheduling strategy as transferring the computing power service executing the first computing power service request to the fifth target computing power node.

9. The method according to claim 8, wherein The method further includes: Allocating computing power resource information to the second target computing power node or the fifth target computing power node.

10. The method according to claim 9, wherein The computing power resource information includes at least one of the following: Computing power service information; Occupied computing power resource information.

11. A computing power scheduling device, characterized in that, It includes: An acquisition unit, configured to acquire resource information reported by at least one computing power node and request information of a target computing power service; A first determination unit, configured to determine a carbon usage efficiency parameter corresponding to each computing power node according to the resource information; A second determination unit, configured to determine a scheduling strategy for the target computing power service based on the request information and the carbon usage efficiency parameter; A scheduling unit, configured to schedule and / or transfer the target computing power service by using the scheduling strategy.

12. A network device, characterized in that, It includes: A processor and a memory for storing a computer program that can run on the processor, wherein, when the processor is used to run the computer program, it executes the steps of the method according to any one of claims 1 to 10.

13. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 10.

Images