Target task allocation method, device and system and electronic equipment
By combining clean energy power, peak-shaving and frequency-modulating power and node resource status in the target task allocation, the target center node is determined, and the problem of insufficient consideration when allocating tasks in the existing technology is solved, and more efficient resource utilization and maximum benefits are achieved change.
Patent Information
- Application Number
- CN202310834371.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-07-07
- Publication Date
- 2025-01-07
AI Technical Summary
When allocating target tasks, the prior art only considers the node resource status, lacks comprehensiveness, and fails to effectively utilize the benefits of clean energy and peak-shaving and frequency-modulating power.
By receiving the target task, the revenue functions of multiple data center nodes are determined, including the consumption of clean energy electricity, auxiliary peak-shaving and frequency-modulation electricity, and the completion of task revenue items. According to these revenue functions and constraints, the clean energy electricity is allocated and the clean energy is calculated. Index, combined with the node resource status, determine the target center node to assign tasks.
With the comprehensive consideration of multiple profit items, it can maximize the consumption of clean energy and the smooth execution of tasks, and improve the efficiency and benefits of task allocation.
Smart Images
Figure CN119271386A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing, and in particular to a target task allocation method, device, system and electronic equipment. Background Art
[0002] When allocating target tasks in related technologies, allocation is performed according to the node resource status, and technical issues with fewer levels are considered.
[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention
[0004] The embodiments of the present invention provide a method, device, system and electronic device for allocating target tasks, so as to at least solve the technical problem that when allocating target tasks in the related art, the target tasks are allocated according to the node resource status, which takes less layers into consideration.
[0005] According to one aspect of an embodiment of the present invention, a target task allocation method is provided, comprising: receiving a target task, wherein the target task carries a target benefit; determining benefit functions corresponding to a plurality of data center nodes respectively, and constraints corresponding to the plurality of benefit functions respectively, based on the target benefit of the target task, wherein each benefit function in the plurality of benefit functions is a function with the maximum benefit as the goal, including a clean energy power consumption benefit item, an auxiliary peak-shaving and frequency-regulating power benefit item, and a task completion benefit item; determining the clean energy consumption power corresponding to the plurality of data center nodes respectively, based on the plurality of benefit functions under the corresponding constraints, wherein the corresponding clean energy consumption power is the power allocated to the corresponding data center node by the clean energy power station; determining the clean energy index corresponding to the plurality of data center nodes respectively, based on the clean energy consumption power corresponding to the plurality of data center nodes respectively; determining the target central node that executes the target task from the plurality of data center nodes according to the clean energy index corresponding to the plurality of data center nodes respectively, and the node resource status corresponding to the plurality of data center nodes respectively; and allocating the target task to the target central node.
[0006] Optionally, determining the clean energy indexes corresponding to the multiple data center nodes respectively based on the clean energy amounts consumed respectively corresponding to the multiple data center nodes includes: determining the clean energy amounts respectively acquired by the multiple data center nodes, and the total demand amounts respectively corresponding to the multiple data center nodes; determining the sum of the clean energy amounts respectively acquired by the multiple data center nodes and the clean energy amounts consumed respectively corresponding to the multiple data center nodes, to obtain the total clean energy amounts respectively corresponding to the multiple data center nodes; determining the ratio of the total clean energy amounts respectively corresponding to the multiple data center nodes to the total demand amounts respectively corresponding to the multiple data center nodes, to obtain the clean energy indexes respectively corresponding to the multiple data center nodes.
[0007] Optionally, determining the target central node that performs the target task from the multiple data center nodes based on the clean energy indexes corresponding to the multiple data center nodes, and the node resource statuses corresponding to the multiple data center nodes, includes: determining a first node set from the multiple data center nodes based on the clean energy indexes corresponding to the multiple data center nodes, and the node resource statuses corresponding to the multiple data center nodes; determining whether there are conflicting nodes in the first node set to obtain a conflict result; when the conflict result is that there are conflicting nodes in the first node set, screening out the conflicting nodes in the first node set to obtain a second node set; and determining the target central node that performs the target task from the second node set.
[0008] Optionally, the determining whether there are conflicting nodes in the first node set and obtaining the conflict result includes: determining the arranged tasks corresponding to multiple data center nodes in the first node set respectively, when the target task carries a target limit time; determining the limit times corresponding to the respectively corresponding arranged tasks; and determining whether there are conflicting nodes in the first node set based on the limit times corresponding to the respectively corresponding arranged tasks to obtain the conflict result.
[0009] Optionally, before determining the benefit functions corresponding to multiple data center nodes respectively and the constraints corresponding to the multiple benefit functions respectively based on the target benefit of the target task, it also includes: determining the server rated power in the corresponding data center node, the server working hours in the corresponding data center node, and the unit electricity price when each benefit function in the multiple benefit functions includes a server operating cost item; determining the server operating cost item based on the server rated power in the corresponding data center node, the server working hours in the corresponding data center node, and the unit electricity price.
[0010] Optionally, before determining the profit functions corresponding to multiple data center nodes respectively and the constraints corresponding to the multiple profit functions respectively based on the target profit of the target task, it also includes: when each of the multiple profit functions includes a server depreciation cost item, determining the server purchase price and the server depreciation rate in the corresponding data center node; determining the server depreciation cost item based on the server purchase price and the server depreciation rate in the corresponding data center node.
[0011] According to one aspect of an embodiment of the present invention, a target task allocation device is provided, comprising: a receiving module for receiving a target task, wherein the target task carries a target benefit; a first determining module for determining, based on the target benefit of the target task, benefit functions corresponding to a plurality of data center nodes, and constraints corresponding to the plurality of benefit functions, wherein each of the plurality of benefit functions is a function with the maximum benefit as the goal, including a clean energy electricity benefit item, an auxiliary peak and frequency regulation electricity benefit item, and a task completion benefit item; a second determining module for determining, based on the plurality of benefit functions under the corresponding constraints, the clean energy electricity benefit items corresponding to the plurality of data center nodes, Energy power, wherein the corresponding absorbed clean energy power is the power allocated by the clean energy power station to the corresponding data center node; a third determination module, used to determine the clean energy indexes corresponding to the multiple data center nodes respectively according to the absorbed clean energy powers corresponding to the multiple data center nodes respectively; a fourth determination module, used to determine the target central node that executes the target task from the multiple data center nodes according to the clean energy indexes corresponding to the multiple data center nodes respectively and the node resource states corresponding to the multiple data center nodes respectively, so as to allocate the target task to the target central node; an allocation module, used to allocate the target task to the target central node.
[0012] According to one aspect of an embodiment of the present invention, a target task allocation system is provided, comprising: a target task allocation device as claimed in claim 7, a data center system, and a clean energy system, wherein the data center system includes multiple data center nodes, and the clean energy system includes a clean energy power generation station.
[0013] According to one aspect of an embodiment of the present invention, there is provided an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement any of the target task allocation methods described above.
[0014] According to one aspect of an embodiment of the present invention, a computer-readable storage medium is provided. When instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute any of the target task allocation methods described above.
[0015] In an embodiment of the present invention, a target task carrying a target benefit is received, and based on the target benefit of the target task, benefit functions corresponding to a plurality of data center nodes and constraints corresponding to the plurality of benefit functions are determined, so that the clean energy consumption corresponding to the plurality of data center nodes can be determined based on the plurality of benefit functions under the corresponding constraints. Based on the clean energy consumption corresponding to the plurality of data center nodes, the clean energy index corresponding to the plurality of data center nodes is determined, and then based on the clean energy index corresponding to the plurality of data center nodes and the node resource status, the target center node for executing the target task is determined from the plurality of data center nodes, so as to achieve the purpose of allocating the target task to the target center node for execution. Since each of the plurality of benefit functions is a function with the maximum benefit as the goal, including the clean energy consumption benefit item, the auxiliary peak and frequency regulation benefit item, and the task completion benefit item, it is possible to obtain the clean energy consumption when the comprehensive benefit of the above-mentioned benefit items is maximized. Furthermore, because the target central node for executing the target task is determined from multiple data center nodes based on the amount of clean energy consumed and the node resource status, the determined target central node takes into account the benefits and its own computing resources, and can ensure that the target task can be smoothly executed when allocated to the target center and can bring more benefits. This solves the technical problem of allocating target tasks in related technologies according to the node resource status, which takes fewer aspects into consideration. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:
[0017] Figure 1 is a flow chart of a target task allocation method according to an embodiment of the present invention;
[0018] Figure 2 It is a schematic diagram of an architecture provided by an optional implementation mode of the present invention;
[0019] Figure 3 4 is a structural block diagram of a target task allocation device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0020] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0021] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0022] Example 1
[0023] According to an embodiment of the present invention, an embodiment of a target task allocation method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0024] Figure 1 is a flow chart of a target task allocation method according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:
[0025] Step S102, receiving a target task, wherein the target task carries a target benefit;
[0026] In step S102 provided in the present application, the target task may be a task that can be executed by a data center node. For example, the target task may be a cloud server resource requested by a user. When a user requests a cloud server resource, a certain fee is required. The fee will be partially or fully allocated to the corresponding data center node, thereby bringing certain benefits. The target benefit is the benefit brought by the corresponding data center node after executing the target task.
[0027] Step S104, according to the target revenue of the target task, determine the revenue functions corresponding to the multiple data center nodes respectively, and the constraints corresponding to the multiple revenue functions respectively, wherein each of the multiple revenue functions is a function with the maximum revenue as the goal, including the revenue item of absorbing clean energy electricity, the revenue item of assisting peak load and frequency regulation electricity, and the revenue item of completing the task;
[0028] In step S104 provided in the present application, the revenue functions corresponding to the multiple data center nodes and the constraints corresponding to the multiple revenue functions can be determined according to the target revenue of the target task. Since the revenue function is a function with the maximum revenue as the goal, the corresponding clean energy electricity collected when the revenue is maximized can be determined through the revenue function.
[0029] Among them, the clean energy electricity consumption income item is the income brought by the data center node consuming clean energy electricity. The reasons are as follows: Clean energy power stations generally supply a predetermined amount of electricity to the power grid. Since clean energy power stations generally generate electricity through clean power generation methods such as wind power, the amount of electricity generated in this way is usually unstable. At some times, the amount of electricity generated by clean power generation may exceed the above predetermined amount. At this time, in order to avoid wasting electricity, the excess electricity can be allocated to the corresponding data center nodes to power the data center nodes, which can bring certain benefits.
[0030] Among them, the auxiliary peak-shaving and frequency-regulating electricity revenue item is for the data center nodes to assist the power grid or clean energy system, helping them to share the revenue brought by behaviors such as electricity consumption and transmission during peak hours.
[0031] The completed task benefit item is the benefit that the above-mentioned data center node can obtain after executing the task.
[0032] It should be noted that the above-mentioned data center node is the central node in the computing power network used to execute tasks. The computing power network usually includes many data center nodes, and the method provided in this application is used to determine which data center node is more appropriate to assign the target task to.
[0033] Step S106, determining the clean energy consumption power corresponding to the multiple data center nodes respectively according to the multiple benefit functions under the corresponding constraint conditions, wherein the corresponding clean energy consumption power is the power allocated by the clean energy power station to the corresponding data center node;
[0034] In step S106 provided in the present application, the clean energy power consumed corresponding to the multiple data center nodes is determined based on multiple profit functions under corresponding constraints. Since the clean energy power consumed represents the power allocated to the corresponding data center node by the clean energy power plant, a portion of the power can be provided to the corresponding data center node, so that the data center node purchases less electricity from the power grid, thereby reducing costs.
[0035] Step S108, determining clean energy indexes corresponding to the multiple data center nodes respectively according to the clean energy consumed by the multiple data center nodes respectively;
[0036] In step S108 provided in the present application, the clean energy indexes corresponding to the multiple data center nodes are determined based on the clean energy power consumed by the multiple data center nodes. Through the clean energy index, it can be determined whether the data center nodes use more or less clean energy, and the impact on the environment, air, etc. can be considered.
[0037] Step S110, determining a target central node for executing a target task from among the multiple data center nodes according to the clean energy indexes respectively corresponding to the multiple data center nodes and the node resource states respectively corresponding to the multiple data center nodes;
[0038] In step S110 provided in the present application, when selecting a data center node, the node's clean energy index and node resource status are considered. By considering these two aspects, the determined target central node can ensure the smooth execution of the target task, and executing target task three can bring more benefits.
[0039] Step S112, assigning the target task to the target central node.
[0040] In step S112 provided in the present application, the purpose of enabling the target central node to execute the target task is achieved.
[0041] Through the above steps S102-S112, a target task with a target benefit is received, and based on the target benefit of the target task, the benefit functions corresponding to the multiple data center nodes and the constraints corresponding to the multiple benefit functions are determined, so that the clean energy consumption corresponding to the multiple data center nodes can be determined based on the multiple benefit functions under the corresponding constraints. Based on the clean energy consumption corresponding to the multiple data center nodes, the clean energy index corresponding to the multiple data center nodes is determined, and then based on the clean energy index corresponding to the multiple data center nodes and the node resource status, the target central node for executing the target task is determined from the multiple data center nodes, so as to achieve the purpose of allocating the target task to the target central node for execution. Since each of the multiple benefit functions is a function with the maximum benefit as the goal, including the clean energy consumption benefit item, the auxiliary peak and frequency regulation power benefit item, and the task completion benefit item, it is possible to obtain the clean energy consumption power when the comprehensive benefit of the above-mentioned benefit items is maximized. Furthermore, because the target central node for executing the target task is determined from multiple data center nodes based on the amount of clean energy consumed and the node resource status, the determined target central node takes into account the benefits and its own computing resources, and can ensure that the target task can be smoothly executed when allocated to the target center and can bring more benefits. This solves the technical problem of allocating target tasks in related technologies according to the node resource status, which takes fewer aspects into consideration.
[0042] As an optional embodiment, based on the clean energy amounts consumed by the multiple data center nodes, respectively, clean energy indexes corresponding to the multiple data center nodes are determined, including: determining the clean energy amounts obtained by the multiple data center nodes, and the total demand amounts corresponding to the multiple data center nodes; determining the sum of the clean energy amounts obtained by the multiple data center nodes and the clean energy amounts consumed by the multiple data center nodes, respectively, to obtain the total clean energy amounts corresponding to the multiple data center nodes; determining the ratio of the total clean energy amounts corresponding to the multiple data center nodes to the total demand amounts corresponding to the multiple data center nodes, respectively, to obtain the clean energy indexes corresponding to the multiple data center nodes.
[0043] In this embodiment, the specific steps for determining the clean energy index are introduced. First of all, it should be noted that since the clean energy electricity consumed is clean energy, and it can power the data center node. However, it should not be enough for the data center node to rely solely on the clean energy electricity consumed. Therefore, it is necessary to determine the total demand electricity of the data center node, subtract the clean energy electricity consumed from the total demand electricity, and obtain the amount of electricity that the data center node should purchase from the power grid. When the data center node purchases electricity from the power grid, it can spontaneously decide whether to purchase non-clean energy electricity such as coal power, or clean energy electricity such as wind power.
[0044] Therefore, the clean energy index of the data center node can be determined by the above method based on the clean energy power purchased by the data center node, the clean energy power consumed, and the total power demand. Among them, Q datacenter Represents the total power demand of the data center node, including the power consumption of computing tasks and other loads; Q clean energy Represents the amount of clean energy purchased from the power grid; QA is the total amount of clean energy consumed. Each of the above values is a measured value within a fixed time period and can be updated in real time.
[0045] Through this step, the clean energy index of clean energy electricity can be accurately calculated.
[0046] As an optional embodiment, a target central node for executing a target task is determined from multiple data center nodes based on clean energy indexes corresponding to the multiple data center nodes, and node resource states corresponding to the multiple data center nodes, including: determining a first node set from multiple data center nodes based on clean energy indexes corresponding to the multiple data center nodes, and node resource states corresponding to the multiple data center nodes; determining whether there are conflicting nodes in the first node set to obtain a conflict result; if the conflict result is that there are conflicting nodes in the first node set, screening out the conflicting nodes in the first node set to obtain a second node set; and determining the target central node for executing the target task from the second node set.
[0047] In this embodiment, the conflict situation is considered in the process of determining the target central node. It should be noted that since the target task may be a variety of types of tasks, the requirements for executing the tasks are also different. For example, some tasks may require execution time, that is, they must be completed within one hour. In this case, it is necessary to consider whether there is a conflict. That is, whether other tasks that need to be processed in a timely manner have been arranged in the data center node, and although the computing power resources are abundant, there is still no way to process them in a timely manner, and the user's requirements cannot be met.
[0048] At this time, the above steps can be used to determine the first node set from the multiple data center nodes according to the clean energy indexes corresponding to the multiple data center nodes and the node resource states corresponding to the multiple data center nodes. In this process, a dual-objective solution algorithm can be used for determination. For example, a dual-objective Pareto solution can be used to obtain a node set that has better benefits and sufficient computing power to perform the target task.
[0049] Then it can be determined whether there are conflicting nodes in the first node set, and the conflict result can be obtained. If the conflict result is that there are conflicting nodes in the first node set, the conflicting nodes in the first node set are screened out to obtain the second node set. That is, if the conflict is the above-mentioned time conflict, the nodes that cannot process the target task in time can be screened out to obtain the second node set, so as to determine the target central node for performing the target task from the second node set. When determining the target central node for performing the target task from the second node set, it can be selected based on the advantages that these nodes can bring to the execution of the task. Through this embodiment, the problem of poor user experience caused by the inability to complete the target task due to the conflict can be avoided.
[0050] As an optional embodiment, determining whether there are conflicting nodes in the first node set and obtaining a conflict result includes: determining, when a target task carries a target limit time, the arranged tasks corresponding to multiple data center nodes in the first node set; determining the limit times corresponding to the respectively corresponding arranged tasks; and determining whether there are conflicting nodes in the first node set based on the limit times corresponding to the respectively corresponding arranged tasks to obtain a conflict result.
[0051] In this embodiment, the case where the conflict is a time limit conflict is introduced. In this case, it can be determined whether the target task can be processed in time, that is, whether there is a conflict result. When the data center node includes one or more arranged tasks, the time limits corresponding to the arranged tasks are determined to see whether the target task can be inserted into the arranged tasks without affecting the execution of other tasks, that is, other tasks also meet their respective time limits. If so, there is no conflict. If not, there is a conflict. Through this embodiment, the problem of poor user experience caused by the inability to complete the target task due to conflicts can be avoided.
[0052] As an optional embodiment, before determining the profit functions corresponding to multiple data center nodes respectively and the constraints corresponding to the multiple profit functions respectively based on the target profit of the target task, it also includes: when each of the multiple profit functions includes a server operating cost item, determining the server rated power in the corresponding data center node, the server working hours in the corresponding data center node, and the unit electricity price; determining the server operating cost item based on the server rated power in the corresponding data center node, the server working hours in the corresponding data center node, and the unit electricity price.
[0053] In this embodiment, the operating cost of the servers included in the data center node is considered in the revenue function. The number of server models in different data center nodes may be different, so it is necessary to calculate according to the specific servers. By considering this aspect, it can be more accurate to consider the maximum revenue based on the revenue function.
[0054] As an optional embodiment, before determining the profit functions corresponding to multiple data center nodes respectively and the constraints corresponding to the multiple profit functions respectively based on the target profit of the target task, it also includes: when each of the multiple profit functions includes a server depreciation cost item, determining the server purchase price and the server depreciation rate in the corresponding data center node; determining the server depreciation cost item based on the server purchase price and the server depreciation rate in the corresponding data center node.
[0055] In this embodiment, the depreciation cost of the servers included in the data center node is considered in the revenue function. The number of server models in different data center nodes may be different, so it is necessary to calculate according to the specific servers. By considering this aspect, it can be more accurate to consider the maximum revenue based on the revenue function.
[0056] Based on the above embodiments and optional embodiments, an optional implementation is provided, which is described in detail below.
[0057] In related technologies, the energy network (the same as the clean energy system) and the computing power network (the same as the data center system) have not been integrated. The computing power network still adopts a computing power-first approach to process computing power tasks. After the integration, there will be a lack of a unified computing task allocation method.
[0058] Based on this, the optional implementation of the present invention considers the integration of energy network and computing network, and proposes an optimal target task allocation method for energy computing network, which not only solves the problem of computing power and energy consumption, but also considers green and low-carbon requirements. The optional implementation of the present invention adds data on the proportion of clean energy, so that the energy computing network center can achieve a balance between computing power, energy and income under comprehensive consideration, and promote the consumption or peak and frequency regulation of green and clean energy. Figure 2 is a schematic diagram of an architecture provided by an optional implementation mode of the present invention, such as Figure 2 As shown, the optional implementation manner of the present invention is introduced below:
[0059] S1, the user puts forward computing requirements on the energy network, collects pending computing tasks, decomposes computing tasks, and obtains multiple target tasks, X = {X1, X2, X3...X n}.
[0060] The following description will be continued by taking a target task as an example.
[0061] S2, the computing network receives the target task, where the target task carries the target revenue;
[0062] S3, according to the target revenue of the target task, determine the revenue functions corresponding to the multiple data center nodes respectively, and the constraints corresponding to the multiple revenue functions respectively, wherein each of the multiple revenue functions is a function with the maximum revenue as the goal, including the revenue item of absorbing clean energy electricity, the revenue item of assisting peak load and frequency regulation electricity, the revenue item of completing the task, the server operation cost item, the server depreciation cost item, and the data center node operation and maintenance cost item;
[0063] Among them, the profit function can be shown as follows:
[0064]
[0065] In the formula, G t is the profit obtained by the data center node at time t; FA t is the income obtained by helping clean energy power generation stations in the energy network to consume clean energy electricity at time t (same as the above-mentioned income item of consuming clean energy electricity); FP t is the income obtained from assisting peak load regulation and frequency regulation at time t (same as the above-mentioned item of electricity income from assisting peak load regulation and frequency regulation); R it is the income obtained by completing task i within time period t (same as the income item of completing task above); S j is the rated power of the jth server; T is the working time of the server; EP t is the current electricity price at time t; λ is the depreciation rate of the server; SP k is the price of the kth server, and C is the fixed cost of running the data center, that is, the operation and maintenance cost (the same as the above data center node operation and maintenance cost item).
[0066] Among them, FA t =α*QA t ;
[0067] In the formula, α is the income of consuming unit clean energy electricity, QA t is the total amount of clean energy consumed at time t.
[0068] The corresponding constraints are as follows:
[0069] QA t <QN t -QS t ;
[0070] Where QN t is the total amount of clean energy generated at time t, QS t It is the total amount of electricity sold by the clean energy station to the grid at time t.
[0071] S4, determining the clean energy consumption power corresponding to the multiple data center nodes respectively according to the multiple benefit functions under the corresponding constraint conditions, wherein the corresponding clean energy consumption power is the power allocated by the clean energy power station to the corresponding data center node;
[0072] Optionally, a single-objective optimization particle swarm algorithm can be used to solve multiple benefit functions under corresponding constraints.
[0073] S5, determining clean energy indexes corresponding to the multiple data center nodes respectively according to the clean energy consumption power corresponding to the multiple data center nodes respectively;
[0074] For example, it is calculated by the following formula: Among them, Q datacenter Represents the total power demand of the data center node, including the power consumption of computing tasks and other loads; Q cleanenergy Represents the amount of clean energy purchased from the power grid; QA is the total amount of clean energy consumed. Each of the above values is a measured value within a fixed time period and can be updated in real time.
[0075] S6, determining a target central node for executing a target task from among the multiple data center nodes according to the clean energy indexes corresponding to the multiple data center nodes and the node resource states corresponding to the multiple data center nodes;
[0076] That is, based on the clean energy indexes corresponding to the multiple data center nodes and the node resource states corresponding to the multiple data center nodes, a comprehensive matching degree calculation for multi-objective optimization is performed for the target tasks and the data center nodes.
[0077] 1) The clean energy index and node resource status corresponding to each data center node are respectively used to perform a dual-objective Pareto solution to form an optimal solution set, where the optimal solution set includes data center nodes with high clean energy index and sufficient computing resources. These data center nodes will be given priority in allocating computing tasks.
[0078] 2) In the optimal solution set, determine whether the tasks executed in the data center node conflict with the target task.
[0079] 4) Match the other information on the information list corresponding to the data center nodes that do not conflict in the optimal solution set with the target task, and filter out the list of computing centers that can handle the target task.
[0080] 5) Determine the target center node from the computing power center list.
[0081] Among them, Table 1 is a schematic diagram of the computing power center list. As shown in Table 1, the target center node can be determined based on the information in the computing power center list.
[0082] Node Number Green Energy Index Number of servers Logical operation amount Matrix operations Tensor operations Storage size Latency metrics 1 2 3 …
[0083] S7, assign the target task to the target central node.
[0084] It should be noted that after the computing tasks are assigned, the remaining computing resources and clean energy index will be dynamically updated.
[0085] Through the above optional implementation, at least the following beneficial effects can be achieved:
[0086] (1) Explore new computing task allocation methods under the integration of energy network and computing power network, and effectively implement the energy computing power network computing task allocation method that takes energy and computing power into consideration;
[0087] (2) Build a profit function, explore the profit function from the perspective of data center nodes and solve it using particle swarm algorithm, solving the problem of optimizing the profit of data center nodes;
[0088] (3) Introducing the clean energy index, the computing network accepts tasks by submitting a list of resource information to the data center node, which then compiles all the information and uploads it to the computing network, which then transmits it to the energy network. The energy network finally allocates computing tasks. Instead of passively accepting all computing tasks, the clean energy index is used as a consideration for data center nodes, so that data center nodes with a high clean energy index can be better allocated to the next task, thereby motivating data center nodes to actively increase the proportion of green clean energy in electricity consumption.
[0089] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.
[0090] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of various embodiments of the present invention.
[0091] Example 2
[0092] According to an embodiment of the present invention, a device for implementing the above target task allocation method is also provided. Figure 3 is a structural block diagram of a target task allocation device according to an embodiment of the present invention. Figure 3 As shown, the device includes: a receiving module 302, a first determining module 304, a second determining module 306, a third determining module 308, a fourth determining module 310 and an allocating module 312. The device is described in detail below.
[0093] A receiving module 302 is used to receive a target task, wherein the target task carries a target profit; a first determining module 304 is connected to the above-mentioned receiving module 302, and is used to determine the profit functions corresponding to the multiple data center nodes respectively, and the constraints corresponding to the multiple profit functions respectively, according to the target profit of the target task, wherein each of the multiple profit functions is a function with the maximum profit as the goal, including a clean energy power consumption profit item, an auxiliary peak-shaving and frequency-regulating power profit item, and a task completion profit item; a second determining module 306 is connected to the above-mentioned first determining module 304, and is used to determine the clean energy power consumption corresponding to the multiple data center nodes respectively, according to the multiple profit functions under the corresponding constraints, wherein the corresponding clean energy power consumption The amount of electricity allocated to the corresponding data center nodes by the clean energy power station; a third determination module 308, connected to the above-mentioned second determination module 306, for determining the clean energy indexes corresponding to the multiple data center nodes respectively according to the clean energy amounts consumed corresponding to the multiple data center nodes respectively; a fourth determination module 310, connected to the above-mentioned third determination module 308, for determining the target central node for executing the target task from the multiple data center nodes according to the clean energy indexes corresponding to the multiple data center nodes respectively and the node resource states corresponding to the multiple data center nodes respectively, so as to allocate the target task to the target central node; an allocation module 312, connected to the above-mentioned fourth determination module 310, for allocating the target task to the target central node.
[0094] It should be noted here that the above-mentioned receiving module 302, first determination module 304, second determination module 306, third determination module 308, fourth determination module 310 and allocation module 312 correspond to steps S102 to S112 in the implementation of the target task allocation method, and the instances and application scenarios implemented by the multiple modules and corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned embodiment 1.
[0095] Example 3
[0096] According to another aspect of an embodiment of the present invention, a target task allocation system is also provided, characterized in that it includes: the target task allocation device as mentioned above, a data center system, and a clean energy system, wherein the data center system includes multiple data center nodes, and the clean energy system includes a clean energy power generation station.
[0097] Example 4
[0098] According to another aspect of an embodiment of the present invention, there is provided an electronic device, comprising: a processor; and a memory for storing processor executable instructions, wherein the processor is configured to execute the instructions to implement any of the above target task allocation methods.
[0099] Example 5
[0100] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. When instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can execute any of the above target task allocation methods.
[0101] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.
[0102] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0103] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.
[0104] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.
[0105] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0106] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.
[0107] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A target task allocation method, characterized in that: include: Receiving a target task, wherein the target task carries a target benefit; According to the target revenue of the target task, determine the revenue functions corresponding to the multiple data center nodes respectively, and the constraints corresponding to the multiple revenue functions respectively, wherein each of the multiple revenue functions is a function with the maximum revenue as the goal, including a revenue item for absorbing clean energy electricity, a revenue item for assisting peak load and frequency regulation electricity, and a revenue item for completing the task; According to the multiple profit functions under the corresponding constraint conditions, determine the clean energy consumption power corresponding to the multiple data center nodes respectively, wherein the corresponding clean energy consumption power is the power allocated by the clean energy power station to the corresponding data center node; Determine the clean energy indexes corresponding to the multiple data center nodes respectively according to the clean energy consumed by the multiple data center nodes respectively; Determining a target central node for executing the target task from among the multiple data center nodes according to the clean energy indexes respectively corresponding to the multiple data center nodes and the node resource states respectively corresponding to the multiple data center nodes; Allocate the target task to the target central node.
2. The method according to claim 1, characterized in that The determining of the clean energy indexes corresponding to the plurality of data center nodes respectively according to the clean energy consumption amounts respectively corresponding to the plurality of data center nodes comprises: Determine the clean energy power respectively obtained by the multiple data center nodes and the total power demand corresponding to the multiple data center nodes; Determine the sum of the clean energy power respectively acquired by the multiple data center nodes and the clean energy power respectively consumed by the multiple data center nodes to obtain the total clean energy power respectively corresponding to the multiple data center nodes; The ratio of the total clean energy power corresponding to the multiple data center nodes to the total demand power corresponding to the multiple data center nodes is determined to obtain the clean energy index corresponding to the multiple data center nodes.
3. The method according to claim 1, characterized in that The step of determining a target central node for executing the target task from the multiple data center nodes based on the clean energy indexes corresponding to the multiple data center nodes and the node resource states corresponding to the multiple data center nodes includes: Determining a first node set from the plurality of data center nodes according to the clean energy indexes respectively corresponding to the plurality of data center nodes and the node resource states respectively corresponding to the plurality of data center nodes; Determine whether there is a conflicting node in the first node set, and obtain a conflict result; When the conflict result is that there are conflicting nodes in the first node set, the conflicting nodes in the first node set are removed to obtain a second node set; A target central node for executing the target task is determined from the second node set.
4. The method according to claim 3, characterized in that The determining whether there is a conflicting node in the first node set to obtain a conflict result includes: In a case where the target task carries a target time limit, determining arranged tasks corresponding to a plurality of data center nodes in the first node set respectively; Determine the time limits for the respective scheduled tasks; According to the respective time limits corresponding to the respectively corresponding arranged tasks, it is determined whether there are conflicting nodes in the first node set to obtain the conflict result.
5. The method according to claim 1, characterized in that Before determining the profit functions corresponding to the plurality of data center nodes and the constraints corresponding to the plurality of profit functions respectively according to the target profit of the target task, the method further includes: In the case where each of the multiple benefit functions includes a server operating cost item, determining a server rated power in a corresponding data center node, a server operating time in a corresponding data center node, and a unit electricity price; The server operation cost item is determined according to the rated power of the server in the corresponding data center node, the working time of the server in the corresponding data center node, and the unit electricity price.
6. The method according to any one of claims 1 to 5, characterized in that Before determining the profit functions corresponding to the plurality of data center nodes and the constraints corresponding to the plurality of profit functions respectively according to the target profit of the target task, the method further includes: In the case where each of the multiple revenue functions includes a server depreciation cost item, determining a purchase price of a server in a corresponding data center node and a server depreciation rate; The server depreciation cost item is determined according to the purchase price of the server in the corresponding data center node and the server depreciation rate.
7. A target task allocation device, characterized in that: include: A receiving module, used for receiving a target task, wherein the target task carries a target benefit; A first determination module is used to determine the profit functions corresponding to the multiple data center nodes respectively and the constraints corresponding to the multiple profit functions respectively according to the target profit of the target task, wherein each of the multiple profit functions is a function with the maximum profit as the goal, including a clean energy electricity profit item, an auxiliary peak and frequency regulation electricity profit item, and a task completion profit item; A second determination module is used to determine the clean energy consumption power corresponding to the multiple data center nodes respectively according to the multiple benefit functions under the corresponding constraint conditions, wherein the corresponding clean energy consumption power is the power allocated by the clean energy power station to the corresponding data center node; A third determination module is used to determine the clean energy indexes corresponding to the multiple data center nodes respectively according to the clean energy consumption power corresponding to the multiple data center nodes respectively; A fourth determination module is used to determine a target central node for executing the target task from the multiple data center nodes according to the clean energy indexes respectively corresponding to the multiple data center nodes and the node resource states respectively corresponding to the multiple data center nodes, so as to allocate the target task to the target central node; The allocation module is used to allocate the target task to the target central node.
8. A target task allocation system, characterized in that: include: The target task allocation device, data center system, and clean energy system as described in claim 7, wherein the data center system includes multiple data center nodes and the clean energy system includes a clean energy power generation station.
9. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the target task allocation method as described in any one of claims 1 to 6.
10. A computer-readable storage medium, characterized in that: When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the target task allocation method as described in any one of claims 1 to 6.