Search

CN-122022960-A - Adaptation method, device, storage medium and processor for computing power leasing

CN122022960ACN 122022960 ACN122022960 ACN 122022960ACN-122022960-A

Abstract

The embodiment of the application provides an adaptation method, equipment, a storage medium and a processor for computing power leasing, and belongs to the technical field of computers. The method comprises the steps of obtaining and analyzing a power leasing task to obtain power leasing parameters, wherein the power leasing parameters comprise operation environment demand parameters, compliance demand parameters, energy consumption demand parameters and user types, determining first power nodes matched with the operation environment demand parameters in available power nodes, determining second power nodes matched with the compliance demand parameters in the first power nodes, determining third power nodes matched with the energy consumption demand parameters in the second power nodes, screening out target power nodes in the third power nodes according to the user types, and distributing the target power nodes to the power leasing task. The scheme provided by the application establishes a multidimensional and self-adaptive dynamic triggering and adjusting mechanism, and meets the requirements of task efficiency, compliance and energy consumption.

Inventors

  • HU ZIXIANG
  • TIAN SHUHANG
  • LIANG SHIXIAN
  • LI HAO

Assignees

  • 中冶京诚数字科技(北京)有限公司
  • 中冶京诚工程技术有限公司

Dates

Publication Date
20260512
Application Date
20251218

Claims (17)

  1. 1. A method of adapting a power lease, the method comprising: Acquiring and analyzing a power leasing task to obtain power leasing parameters, wherein the power leasing parameters comprise an operation environment demand parameter, a compliance demand parameter, an energy consumption demand parameter and a user type; determining a first computing power node matched with the running environment demand parameters in available computing power nodes; determining a second computing force node matched with the compliance requirement parameter in the first computing force node; determining a third computing power node matched with the energy consumption demand parameter in the second computing power nodes; Screening out a target computing node from the third computing nodes according to the user type; And distributing the target computing power node to the computing power leasing task.
  2. 2. The method of adapting a power lease of claim 1, characterized in that among available power nodes, the step of determining a first power node that matches said operating environment demand parameter comprises: Determining an operation environment configuration parameter matched with the operation environment demand parameter by using a first mapping library; and determining the first computing force node matched with the running environment configuration parameters in available computing force nodes by using a fourth mapping library.
  3. 3. The method of adapting a power lease according to claim 2, characterized in that in said first power node, the step of determining a second power node matching said compliance demand parameter comprises: determining a compliance configuration parameter matched with the compliance requirement parameter by using a second mapping library; And determining the second computing force node matched with the compliance configuration parameters in the first computing force node by using the fourth mapping library.
  4. 4. The method of adapting a power lease according to claim 1, characterized in that in said second power node, the step of determining a third power node that matches said energy consumption demand parameter comprises: calculating the energy consumption adaptation degree of the second computing power node by using a third mapping library; And comparing the value of the energy consumption adaptation degree of the second computing node with the energy consumption demand parameter to obtain a third computing node matched with the energy consumption demand parameter.
  5. 5. The method of adapting a power lease of claim 4, wherein said power consumption adaptation is calculated by: ; Wherein, the method comprises the steps of, Representing the energy consumption adaptation degree; Representing a renewable energy duty cycle; representing a renewable energy weight coefficient; representing the relative efficacy ratio; representing the efficiency weight coefficient; representing the unit calculation power consumption optimizing rate; representing a calculation power energy consumption weight coefficient; representing a task energy consumption sensitivity coefficient; Representing the dynamic energy consumption adjustment factor.
  6. 6. The method of adapting a computer program product according to claim 5, wherein: The relative efficacy ratio The calculation formula of (2) is as follows: = Wherein, the method comprises the steps of, Representing the current efficiency ratio of the computing force node; Representing the efficiency ratio of the force node reference; the unit calculation power consumption optimizing rate The calculation formula of (2) is as follows: = Wherein, the method comprises the steps of, Representing the current unit calculation power energy consumption of the calculation power node; representing a power calculation energy consumption reference value of a power calculation node unit; The dynamic energy consumption regulating factor The calculation formula of (2) is as follows: =1- Wherein, the method comprises the steps of, The standard deviation of unit calculation force energy consumption in the preset time of the calculation force node is represented; and the average value of unit calculation force energy consumption in the preset time of the calculation force node is represented.
  7. 7. The method of adapting a power lease of claim 5, characterized in that, before calculating an energy consumption adaptation degree of said second power node using said third mapping library, said method further comprises: acquiring energy consumption feedback data of the completed power leasing task; And adjusting the renewable energy source weight coefficient, the efficiency weight coefficient and the calculated energy consumption weight coefficient according to the energy consumption feedback data.
  8. 8. The method of adapting a power lease of claim 4, wherein said power consumption adaptation is calculated by: Wherein, the method comprises the steps of, Representing the energy consumption adaptation degree; Representing a renewable energy duty cycle; representing the current efficiency ratio of the computing force node; representing the lowest energy efficiency ratio of the industry to which the computing force node belongs; representing the task energy consumption sensitivity coefficient.
  9. 9. The method of adapting a power lease according to claim 1, characterized in that in said second power node, the step of determining a third power node that matches said energy consumption demand parameter comprises: calculating the energy consumption adaptation degree of the second computing power node by using a neural network model; And comparing the value of the energy consumption adaptation degree of the second computing node with the energy consumption demand parameter to obtain a third computing node matched with the energy consumption demand parameter.
  10. 10. The method of adapting a power lease according to claim 1, characterized in that said step of screening out a target power node among said third power nodes according to said user type comprises: Determining whether the user type is a first type; If the user type is the first type, the target computing node is selected from the third computing nodes according to the corresponding energy consumption adaptation degree, the corresponding running environment matching degree, the corresponding typical task computing duration and the corresponding data magnitude; and if the user type is not the first type, screening the target computing node from the third computing node according to the idle period, the typical task computing duration and the data magnitude.
  11. 11. The method of adapting a computing power lease of claim 1, wherein after assigning the target computing power node to the computing power lease task, the method further comprises: monitoring the operation condition of computing resources of a target computing node; If the computing resource running condition of the target computing power node is not matched with the running environment demand parameter, the first computing power node is determined again; If the computing resource running condition of the target computing power node is not matched with the compliance requirement parameter, the second computing power node is determined again; and if the computing resource running condition of the target computing power node is not matched with the energy consumption requirement parameter, the third computing power node is redetermined.
  12. 12. The method for adapting a power lease according to claim 11, wherein said power lease parameters further comprise a cost demand parameter, and wherein after monitoring a computing resource operation condition of a target power node, said method further comprises: and if the resource cost of the target computing power node is not matched with the cost requirement parameter, reconfirming the first computing power node.
  13. 13. The method of adapting a computing power lease of claim 1, wherein after assigning the target computing power node to the computing power lease task, the method further comprises: monitoring the execution condition of the power leasing task; generating an audit log according to the execution condition of the power leasing task; and setting the period for reserving the audit log according to the compliance demand parameters.
  14. 14. The method of adapting a power rental of claim 13, wherein the power rental parameters further comprise a task delay threshold, wherein the performance of the power rental task comprises a delay value for the power rental task, and wherein after assigning the target power node to the power rental task, the method further comprises: and if the delay value of the power leasing task is greater than or equal to the task delay threshold, reconfirming the first power node.
  15. 15. The power leasing adapting device is characterized by comprising a processor and a memory, wherein the memory stores instructions; the processor invokes the instructions in the memory to cause the computing power leased adaptation device to implement the computing power leased adaptation method of any one of claims 1 to 14.
  16. 16. A computer readable storage medium having instructions stored thereon, which when executed by a processor implement the method of adapting a computer-force lease of any one of claims 1 to 14.
  17. 17. A computer program product comprising a computer program which, when executed by a processor, implements the computing power lease adaptation method of any one of claims 1 to 14.

Description

Adaptation method, device, storage medium and processor for computing power leasing Technical Field The invention relates to the technical field of computers, in particular to an adaptation method, equipment, a storage medium and a processor for computing power leasing. Background Computing power leasing is a business model of leasing computing resources through cloud computing service providers, where users can acquire computing power on demand and pay for actual usage without the need for a self-built infrastructure. With the development of artificial intelligence and digital economy, the computing power has become a core productivity, and the computing power leasing market is rapidly increased. However, with the improvement of energy consumption requirement, data compliance requirement and calculation efficiency, the existing power lease scheme cannot simultaneously consider the data compliance requirement, the energy consumption requirement and the efficiency requirement, and particularly for the highly sensitive industries such as medical treatment, finance and the like, the compliance cost during power lease is increased sharply, and meanwhile, the power node cannot be adjusted according to the characteristics of lease tasks and the real-time energy condition, so that the development of green power is restricted. Therefore, how to design an adaptation method for power leasing, which takes into consideration the compliance requirement, the energy consumption requirement and the efficiency requirement, becomes a problem to be solved in the field. Disclosure of Invention In a first aspect, an embodiment of the present invention provides an adapting method for computing power leasing, so as to consider task efficiency requirements, compliance requirements and energy consumption requirements during adaptation, where the method includes: Acquiring and analyzing a power leasing task to obtain power leasing parameters, wherein the power leasing parameters comprise an operation environment demand parameter, a compliance demand parameter, an energy consumption demand parameter and a user type; determining a first computing power node matched with the running environment demand parameters in available computing power nodes; determining a second computing force node matched with the compliance requirement parameter in the first computing force node; determining a third computing power node matched with the energy consumption demand parameter in the second computing power nodes; Screening out a target computing node from the third computing nodes according to the user type; And distributing the target computing power node to the computing power leasing task. Optionally, the step of determining a first computing power node matched with the running environment demand parameter among available computing power nodes includes: Determining an operation environment configuration parameter matched with the operation environment demand parameter by using a first mapping library; and determining the first computing force node matched with the running environment configuration parameters in available computing force nodes by using a fourth mapping library. Further optionally, in the first computing force node, the step of determining a second computing force node that matches the compliance requirement parameter includes: determining a compliance configuration parameter matched with the compliance requirement parameter by using a second mapping library; And determining the second computing force node matched with the compliance configuration parameters in the first computing force node by using the fourth mapping library. Optionally, in the second computing node, the step of determining a third computing node that matches the energy consumption requirement parameter includes: calculating the energy consumption adaptation degree of the second computing power node by using a third mapping library; And comparing the value of the energy consumption adaptation degree of the second computing node with the energy consumption demand parameter to obtain a third computing node matched with the energy consumption demand parameter. Further optionally, the calculation formula of the energy consumption adaptation degree is: ; Wherein, the method comprises the steps of, Representing the energy consumption adaptation degree; Representing a renewable energy duty cycle; representing a renewable energy weight coefficient; representing the relative efficacy ratio; representing the efficiency weight coefficient; representing the unit calculation power consumption optimizing rate; representing a calculation power energy consumption weight coefficient; representing a task energy consumption sensitivity coefficient; Representing the dynamic energy consumption adjustment factor. Further optionally: The relative efficacy ratio The calculation formula of (2) is as follows: = Wherein, the method comprises the steps of, Representing the current efficien