Computing power network node evaluation and operation method and apparatus
View Patent ↗A computing power network node evaluation and operation method includes: determining overall computing power information of a computing power network; evaluating a computing power service for a computing power service demand based on the overall computing power information and the computing power service demand so as to obtain a computing power evaluation value of each computing power node; determining a computing power configuration scheme corresponding to the computing power service demand based on the computing power evaluation value of each computing power node and a computing power calling mode of each computing power node within a time period corresponding to the computing power service demand; and respectively sending computing power operation instructions to a computing power demand node and a computer power service node indicated in the computing power configuration scheme to establish a computing power calling connection, thus completing the computing power service demand.
1 . A computing power network node evaluation and operation method, wherein the method is applied to a computing power server in a computing power network, the computing power network comprises at least one computing power node, and the method comprises the following steps: determining an overall computing power information of the computing power network, wherein the overall computing power information comprises a computing power node information of each computing power node, and a connection relationship between each computing power node;
generating computing power service feature parameters based on the overall computing power information, extracting a quantitative feature served by each computing power service node from the computing power service feature parameters, so as to obtain each computing power service quantitative feature, and determining a computing power service quantitative feature library according to each computing power service quantitative feature;
determining a computing power service feature parameter demand and an algorithm information corresponding to a computing power service demand sent by a computing power demand node, and determining first candidate computing power nodes conforming to the computing power service feature parameter demand and the algorithm information from the computing power service quantitative feature library;
generating a computing power network matrix for the computing power service demand based on a historic computing power service volume between the computing power demand node and each of the first candidate computing power nodes as well as a connection relationship between the computing power demand node and each of the first candidate computing power nodes, wherein the historic computing power service volume is determined based on the overall computing power information; solving the computing power network matrix to obtain a computing power evaluation value of each of the first candidate computing power nodes;
determining a computing power configuration scheme corresponding to the computing power service demand based on the computing power evaluation value of each computing power node and a computing power calling mode of each computing power node within a time period corresponding to the computing power service demand, wherein the computing power calling mode is a self-sufficient mode or an open mode; and
respectively sending computing power operation instructions to a computing power demand node and a computer power service node indicated in the computing power configuration scheme to make the computing power demand node and the computing power service node establish a computing power calling connection, thus completing the computing power service demand.
2 . The computing power network node evaluation and operation method according to claim 1 , wherein the determining a computing power configuration scheme corresponding to the computing power service demand based on the computing power evaluation value of each computing power node and a computing power calling mode of each computing power node within a time period corresponding to the computing power service demand comprises the following steps:
determining a second candidate computing power node from the first candidate computing power nodes, wherein the second candidate computing power node is a computing power node in which the computing calling mode within the time period corresponding to the computing power service demand is an open mode;
determining an optimal computing power path from the second candidate computing power node with at least one of the highest sum of computing power evaluation values of the computing power nodes in a computing power path, the lowest sum of computing power prices of the computing power nodes in the computing power path, and the highest ratio of the sum of the computing power evaluation values to the sum of computing power prices of the computing power nodes in the computing power path as a target; and determining a computing power configuration scheme corresponding to the computing power service demand based on the optimal computing power path;
wherein the computing power price is determined based on the computing power node information.
3 . The computing power network node evaluation and operation method according to claim 1 , wherein after the respectively sending computing power operation instructions to a computing power demand node and a computer power service node indicated in the computing power configuration scheme to make the computing power demand node and the computing power service node establish a computing power calling connection to complete the computing power service demand, the method further comprises the following steps:
supervising the computing power demand node to pay computing power fee to the computing power service node, and receiving computing power state update information returned by the computing power demand node; and
updating the overall computing power information based on the computing power state update information.
4 . The computing power network node evaluation and operation method according to claim 1 , wherein the computing power calling mode of each computing power node is determined by determining whether the computing power service is surplus or not by any computing power node for any computing power service demand of the node;
in a case that the computing power service is surplus, any computing power node determines that the computing power calling mode is an open mode; and in a case that the computing power service is not surplus, the computing power calling mode is the self-sufficient mode.
5 . An electronic device or a quantum device, comprising a memory, a processor, and a computer program stored on the memory and capable of being operated on the processor, wherein the processor, when executing the computer program, is able to implement the computing power network node evaluation and operation method according to claim 1 .
6 . A non-transient computer-readable storage medium, wherein an electronic computer program and/or a quantum computer program are/is stored on the non-transient computer-readable storage medium, and the computer program, when executed by a processor, is able to implement the computing power network node evaluation and operation method according to claim 1 .