Computing power resource pricing method and system oriented to supercomputing internet and computer equipment

Abstract

The invention relates to the technical field of supercomputing internet, in particular to a supercomputing internet-oriented computing power resource pricing method and system and computer equipment, and the method comprises the steps: obtaining unit operation time variable cost and annual fixed cost based on a supercomputing center operation record, and carrying out the conversion to obtain an annual fixed cost depreciation amount; based on the historical computing power resource price mean value and the historical operation time data, obtaining estimated annual operation time through a cubic spline interpolation method; obtaining an estimated computing power resource unit cost based on the unit operation time variable cost, the annual fixed cost depreciation amount and the estimated annual operation time; constructing a hierarchical structure model based on an analytic hierarchy process, and obtaining pricing influence factor weights; obtaining a pre-estimated computing power resource price based on the pre-estimated computing power resource unit cost and the pricing influence factor weight; and performing difference value verification on the pre-estimated computing power resource price and the historical computing power resource price mean value until a difference value meets a preset threshold value, and obtaining a final computing power resource price.

Description

Technical Field

[0001] The present invention relates to the technical field of supercomputer Internet. More specifically, the present invention relates to a computing power resource pricing method, system and computer device for the supercomputer Internet. Background Art

[0002] The supercomputer Internet is an advanced computing infrastructure based on supercomputers and high-speed Internet. It connects supercomputer centers distributed in various places through high-speed networks, aggregates the software and hardware resources of multiple supercomputer centers, and provides users with diversified computing power services. However, with the continuous development of the construction of the supercomputer Internet, the resources such as computing power, storage, and equipment included in the supercomputer Internet vary greatly, and factors such as the operating costs and policy supports of supercomputer centers in different regions are different.

[0003] Currently, traditional computing power resource pricing strategies are mainly based on rough evaluations of past data, and there are many problems: only applicable to a single supercomputer center, the pricing standards are difficult to unify, and the universality is poor; the pricing factors are not comprehensively considered, and only single-dimensional indicators are used to evaluate the computing power resources of service equipment, and the reliability of the evaluation results is low, affecting the cost recovery of the supercomputer center; the relationship between the operating time and the computing power price is ignored, resulting in large deviations in the calculation results, affecting the rationality of the computing power resource pricing, and it is difficult to ensure the normal operation of the supercomputer center.

[0004] Therefore, providing a computing power resource pricing method, system and computer device for the supercomputer Internet with multi-dimensional evaluation, strong universality and reasonable pricing is an urgent problem to be solved at present. Summary of the Invention

[0005] The present invention provides a computing power resource pricing method, system and computer device for the supercomputer Internet to at least solve the problems of single pricing factor and poor reliability of the computing power pricing result in the existing computing power resource pricing strategy.

[0006] To achieve the above object, the present invention provides a computing power resource pricing method for the supercomputer Internet, including: The step of calculating the fixed cost of computing power resources: obtaining the variable cost per unit of operating time and the annual fixed cost based on the operation records of the supercomputer center, and converting the annual fixed cost to obtain the annual fixed cost depreciation amount; The step of calculating the unit cost of computing power resources: obtaining the estimated annual operating time through cubic spline interpolation based on the historical average computing power resource price and historical operating time data; Obtaining the estimated unit cost of computing power resources based on the variable cost per unit of operating time, the annual fixed cost depreciation amount and the estimated annual operating time; Steps for obtaining the weights of influencing factors for computing power resource pricing: Construct a hierarchical structure model based on the analytic hierarchy process, and obtain the weights of pricing influencing factors based on the hierarchical structure model; Steps for estimating computing power resource pricing: Obtain the pre-estimated computing power resource pricing based on the pre-estimated unit cost of computing power resources and the weights of pricing influencing factors; Perform a difference check on the pre-estimated computing power resource pricing and the average historical computing power resource price until the difference meets a preset threshold to obtain the final computing power resource pricing.

[0007] Furthermore, the steps for calculating the unit cost of computing power resources include: Define as the computing power resource price and the running time the cubic spline interpolation function between them. Divide into sample points on the interval ; The cubic spline interpolation function needs to meet the following conditions: The interpolation condition is: ; The continuity condition is: at the interval endpoints , is continuous and the first derivative , the second derivative are continuous; The boundary condition is: and ; Based on the interpolation condition, the continuity condition, and the boundary condition, establish a cubic spline equation system and solve to obtain the cubic spline interpolation function ; Based on the average historical price Pp, obtain the estimated annual running time .

[0008] Furthermore, the steps for calculating the unit cost of computing power resources further include: Based on the annual fixed cost depreciation amount and the estimated annual running time, calculate the fixed cost per unit running time, and the expression is as follows:

[0009] where is the fixed cost per unit running time, is the annual fixed cost depreciation amount, is the estimated annual running time; Based on the fixed cost per unit running time and the variable cost per unit running time, calculate the unit cost of computing power resources, and the expression is as follows:

[0010] Among them, is the unit cost of computing power resources, is the fixed cost per unit of running time, is the variable cost per unit of running time.

[0011] Furthermore, the steps for obtaining the weights of the influencing factors of the computing power resource pricing include: Steps for establishing a hierarchical structure model: Construct the hierarchical structure model, which includes a computing power pricing target layer and a criterion layer connected in sequence. The criterion layer has multiple pricing influencing factors, including: scenario factors, chip factors, storage factors, network factors, computing factors, and device factors; Steps for constructing a comparison matrix: Based on the 1-9 scale method, pairwise compare the multiple pricing influencing factors in the criterion layer to construct a criterion layer comparison matrix; Steps for calculating weights: Based on the geometric mean method, calculate the weights of the pricing influencing factors through the criterion layer comparison matrix, expressed as:

[0012] Among them, is the weight of the pricing influencing factor, is the criterion layer comparison matrix, and .

[0013] Furthermore, the steps for obtaining the weights of the influencing factors of the computing power resource pricing further include: Conduct a consistency test on the criterion layer comparison matrix; Consistency index is:

[0014] Among them, is the consistency index, is the maximum eigenvalue of the criterion layer comparison matrix, is the matrix order; Random consistency index is:

[0015] Among them, is the number of random criterion layer comparison matrices, is the matrix order; Consistency ratio is:

[0016] When the consistency ratio is met, the consistency test is passed; otherwise, reconstruct the criterion layer comparison matrix.

[0017] Further, the steps for estimating the computing power resource pricing include: Obtaining the pre - estimated computing power resource pricing based on the pre - estimated unit cost of the computing power resource and the weight of the pricing influencing factors. The expression is as follows:

[0018]

[0019] Wherein, is the pre - estimated computing power resource pricing, is the profit, is the pre - estimated unit cost of the computing power resource, is the computing power market profit rate, is the user configuration factor parameter, is the weight of the pricing influencing factors.

[0020] Further, the steps for estimating the computing power resource pricing further include: Performing a difference check between the pre - estimated computing power resource pricing and the average value of the historical computing power resource prices. The difference check expression is as follows:

[0021] Wherein, is the preset threshold. When , the pre - estimated computing power resource pricing is reasonable; when , then the pre - estimated computing power resource pricing replaces the average historical price , recalculates the pre - estimated unit cost of the computing power resource , and iterates until the difference meets the preset threshold to obtain the final computing power resource pricing.

[0022] The present invention provides a computing power resource pricing system for a supercomputer internet, which is applied to the above - mentioned computing power resource pricing method for a supercomputer internet, and includes: A computing power resource fixed - cost accounting module: obtaining the variable cost per unit of operation time and the annual fixed cost based on the operation records of the supercomputer center, and performing conversion on the annual fixed cost to obtain the annual fixed - cost depreciation amount; A computing power resource unit - cost accounting module: obtaining the pre - estimated annual operation time through cubic spline interpolation based on the average historical computing power resource price and the historical operation time data; Obtaining the pre - estimated unit cost of the computing power resource based on the variable cost per unit of operation time, the annual fixed - cost depreciation amount, and the pre - estimated annual operation time; A computing power resource pricing influencing - factor weight obtaining module: constructing a hierarchical structure model based on the analytic hierarchy process, and obtaining the weight of the pricing influencing factors based on the hierarchical structure model; Computing power resource pricing estimation module: Obtain the pre - estimated computing power resource pricing based on the pre - estimated unit cost of computing power resources and the weight of pricing influencing factors; Perform a difference check on the pre - estimated computing power resource pricing and the average value of historical computing power resource prices until the difference meets a preset threshold, and obtain the final computing power resource pricing.

[0023] The present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. It is characterized in that when the processor executes the computer program, it implements the computing power resource pricing method for the supercomputer internet as described above.

[0024] The present invention provides a computer - readable storage medium, on which a computer program is stored. It is characterized in that when the program is executed by a processor, it implements the computing power resource pricing method for the supercomputer internet as described above.

[0025] Compared with the prior art, the advantages and positive effects of the present invention are as follows: The present invention comprehensively considers the influence of multiple dimensions such as chips, storage, network, and devices on the pricing of computing power resources, breaking through the limitation of traditional pricing strategies that are only applicable to single supercomputer centers. The present invention fully considers the differences in resource characteristics and operating conditions of each supercomputer center in the supercomputer internet, and can provide a precise and unified resource pricing scheme for the collaborative operation of multiple supercomputer centers, effectively meeting the diverse pricing needs in the complex environment of the supercomputer internet, greatly improving the universality of the pricing strategy, and strongly promoting the reasonable allocation and efficient utilization of supercomputer resources in a wider range.

[0026] The present invention establishes the connection between running time and pricing based on the cubic spline interpolation method, reduces the influence of running time on the uneven allocation of fixed costs, thereby making the cost accounting of computing power resources more accurate and effectively improving the reliability of the computing power pricing result.

[0027] The present invention constructs a computing power resource pricing model based on the analytic hierarchy process, and dynamically adjusts the computing power resource price according to the variable cost fluctuation, greatly improving the rationality and market adaptability of the computing power resource pricing, ensuring the sustainable development of the supercomputer center while improving user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic flowchart of the computing power resource pricing method for the supercomputer internet in the embodiment of the present invention; Figure 2 It is a schematic diagram of the cubic spline interpolation method in the embodiment of the present invention; Figure 3 It is a schematic flowchart of the steps for obtaining the weight of computing power resource pricing influencing factors in the embodiment of the present invention; Figure 4Schematic diagram of the hierarchical structure model in the embodiments of the present invention; Figure 5 Schematic diagram of the 1-9 scale method in the embodiments of the present invention; Figure 6 Schematic diagram of the annotation of the comparison matrix of the criterion layer in the embodiments of the present invention; Figure 7 Schematic diagram of the estimated calculation power resource pricing in the embodiments of the present invention; Figure 8 Schematic diagram of the structure of the calculation power resource pricing system for the supercomputer Internet in the embodiments of the present invention; Figure 9 Schematic diagram of the computer device provided in the embodiments of the present invention.

[0029] In the above figures: 40. Bus; 41. Processor; 42. Memory; 43. Communication interface; 100. Fixed cost accounting module for computing power resources; 200. Unit cost accounting module for computing power resources; 300. Weight acquisition module for influencing factors of computing power resource pricing; 400. Estimated computing power resource pricing module. Specific embodiments

[0030] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0031] In the description of the present application, it should be understood that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation to the present application. The terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise specified, the meaning of "plurality" is two or more.

[0032] In the description of this application, it should be noted that unless otherwise clearly specified and limited, the terms "installation", "connection", and "linkage" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0033] As Figures 1-9 shown, the present invention provides a computing power resource pricing method, system and computer device for the supercomputing Internet, so as to at least solve the problems of single pricing factors and poor reliability of computing power pricing results in the existing computing power resource pricing strategies. The following will be Figures 1-9 described in detail the computing power resource pricing method, system and computer device for the supercomputing Internet provided by the present invention.

[0034] Embodiment 1: Figure 1 is a schematic flowchart of the computing power resource pricing method for the supercomputing Internet in an embodiment of the present invention. As Figure 1 shown, the present invention provides a computing power resource pricing method for the supercomputing Internet, including a computing power resource fixed cost accounting step S100, a computing power resource unit cost accounting step S200, a computing power resource pricing influencing factor weight obtaining step S300, and a computing power resource pricing estimation step.

[0035] Computing power resource fixed cost accounting step S100: Obtain the variable cost per unit of running time and the annual fixed cost based on the operation records of the supercomputing center, and calculate the depreciation amount of the annual fixed cost.

[0036] In some embodiments, the powerful computing power of the supercomputing center comes from its large scale and advanced hardware configuration. Fully considering the cost inputs of various aspects, the costs are divided into two parts: variable costs and fixed costs. Among them, the annual variable costs do not need to be accounted for, and the variable cost per unit of running time can be directly obtained based on the operation records of the supercomputing center. The depreciation amount of the annual fixed cost needs to be calculated according to expenses such as equipment procurement and software transplantation. The calculation method can adopt the declining balance method or the sum-of-the-years'-digits method on an annual basis.

[0037] In some embodiments, the declining balance method is used to calculate the annual fixed cost. The declining balance method is a method of accelerated depreciation. Its core principle is that the depreciation expenses accrued in the early stage are relatively large, and gradually decrease in the later stage. The advantage of this method is that it can fully consider the rapid replacement of fixed assets such as equipment and the depreciation of technology in the early stage of asset use, so that the supercomputing center can recover more costs in the early stage of equipment use. Specifically, the declining balance method is used to calculate the annual fixed cost depreciation amount, and the expression is as follows:

[0038]

[0039] in, is the annual depreciation rate, is the estimated period of fixed cost discount, For the Annual fixed cost residual value, For the This method does not take into account the residual value of the fixed assets purchased at the fixed cost, and the remaining fixed costs should be amortized evenly within two years before the expiration of the fixed cost depreciation period.

[0040] In some embodiments, the annual sum method is used to calculate the annual fixed cost. The annual sum method is a method of calculating the annual depreciation amount based on the year-by-year decrease in the remaining useful life of fixed assets. Compared with other depreciation methods such as the straight-line depreciation method, the annual sum method accrues more depreciation in the early stage and less in the later stage, which can better reflect the value consumption characteristics of assets in different stages of use. It is especially suitable for the depreciation calculation of supercomputing center equipment with rapid technological updates. Specifically, the declining balance method is used to calculate the annual fixed cost depreciation amount, and the expression is as follows:

[0041]

[0042] in, For the Annual depreciation rate, For the The number of years that the annual fixed costs need to be converted is: is the total number of years to be converted for each year, is the estimated period of fixed cost discount, For the Years converted to years, For the Annual fixed cost depreciation, is the total fixed cost, is the estimated net salvage value. At the end of the depreciable years calculated by this method, the sum of the depreciation rates is 1, ensuring the full recovery of the fixed costs of the supercomputing center.

[0043] Calculation steps S200 of the unit cost of computing power resources: Based on the historical average computing power resource price and historical operation time data, obtain the estimated annual operation time through cubic spline interpolation; Obtain the estimated unit cost of computing power resources based on the variable cost per unit operation time, the annual fixed cost depreciation amount, and the estimated annual operation time.

[0044] Steps S300 for obtaining the weights of factors affecting the pricing of computing power resources: Construct a hierarchical structure model based on the analytic hierarchy process, and obtain the weights of factors affecting the pricing based on the hierarchical structure model.

[0045] Steps S400 for estimating the pricing of computing power resources: Obtain the estimated pricing of computing power resources based on the estimated unit cost of computing power resources and the weights of factors affecting the pricing; Perform a difference check on the estimated pricing of computing power resources and the historical average computing power resource price until the difference meets the preset threshold to obtain the final pricing of computing power resources.

[0046] Preferably, as Figure 2 shown, the calculation steps S200 of the unit cost of computing power resources include: Define as the cubic spline interpolation function between the computing power resource price and the operation time . Divide into sample points on the interval , with two endpoints . The cubic spline interpolation function

[0047] needs to satisfy the following interpolation conditions, continuity conditions, and boundary conditions.

[0048] The interpolation conditions ensure that the function value of the cubic spline interpolation function at each sample point is equal to the actual operation time , enabling the constructed function to accurately fit the known data points and providing a reliable basis for subsequent operation time estimation.

[0049] The continuity conditions are: At the interval endpoints , is continuous and the first derivative , the second derivative are continuous, specifically as follows: Continuity conditions:

[0050] First derivative Continuity condition:

[0051] Second derivative Continuity condition:

[0052] The continuity condition ensures that the function has a smooth transition over the entire interval . Taking two adjacent intervals and as an example, at the endpoint , not only the function values should be equal, but also its first derivative and second derivative should be equal. This means that there will be no sudden jumps or turns at the connection points of different intervals of the function, ensuring that the estimated running time changes based on this function are continuous and smooth, which is more in line with the actual situation.

[0053] The boundary conditions are: and

[0054] The boundary conditions further constrain the characteristics of the function at the interval endpoints. and indicate that at the starting and ending points of the interval, the second derivative of the function is 0, that is, the degree of bending of the function at the endpoints is 0, making the function smoother at the boundaries. This helps to improve the stability and accuracy of the function, avoid unreasonable fluctuations at the boundaries, and thus improve the reliability of the running time estimation.

[0055] Based on the above interpolation conditions, continuity conditions, and boundary conditions, a cubic spline equation system is established, solved to obtain the coefficients of the function , and the specific expression of the function is determined.

[0056] The cubic spline equation system is specifically as follows:

[0057]

[0058]

[0059]

[0060]

[0061] Based on the historical price average Pp , obtain the estimated annual running time . The historical price average Pp is the average price of computing power resources in previous years. When using the historical price average for the first time Pp to calculate the estimated annual running time, it can provide a stable and representative iterative starting reference point for the estimated annual running time.

[0062] Preferably, based on the annual fixed cost depreciation amount and the estimated annual running time , calculate the fixed cost per unit running time , and the expression is as follows: Based on the fixed cost per unit running time and the variable cost per unit running time , calculate the unit cost of computing power resources , and the expression is as follows:

[0063] By establishing a non - linear correlation between the computing power resource price and the running time through cubic spline interpolation, it significantly reduces the rationality deviation of the computing power resource pricing caused by the traditional method's neglect of the relationship between the running time and the computing power price.

[0064] Preferably, as Figures 3-6 shown, the steps S300 for obtaining the weights of the influencing factors of computing power resource pricing include the steps of establishing a hierarchical structure model, constructing a comparison matrix, calculating weights, and consistency test steps.

[0065] As Figure 4 shown, the steps of establishing a hierarchical structure model include constructing a hierarchical structure model, and the hierarchical structure model includes a computing power pricing target layer and a criterion layer connected in sequence. The criterion layer has multiple pricing influencing factors, including: scenario factors, chip factors, storage factors, network factors, computing factors, and equipment factors.

[0066] In some embodiments, the computing power pricing target layer is the decision - making purpose, that is, the decision - making purpose is computing power pricing, and the criterion layer is There are several pricing influencing factors, which include but are not limited to the factors listed above, and can be flexibly expanded and adjusted according to the actual operation situation of the supercomputing center, changes in the market environment, and specific user needs. For example, with the continuous development of supercomputing technology and the increasing popularity of the concept of environmental protection, energy efficiency has also become one of the key factors affecting the pricing of computing power, and energy efficiency can be added to the pricing influencing factors.

[0067] As Figures 5-6 shown, the steps of constructing the comparison matrix include pairwise comparing multiple pricing influencing factors in the criterion layer based on the 1-9 scale method to construct the criterion layer comparison matrix.

[0068] In some embodiments, to construct the criterion layer comparison matrix and minimize the difficulty of comparing factors with different natures, pairwise compare the influencing factors in the criterion layer, refer to the 1-9 scale method as Figure 5 shown, and obtain the criterion layer comparison matrix annotation diagram through annotation. As Figure 6 shown, to represent the relative importance of each pricing influencing factor to the pricing of computing power, the expression is as follows:

[0069] The steps of calculating the weight include calculating the weight of the pricing influencing factors based on the geometric mean method through the criterion layer comparison matrix, expressed as:

[0070] where is the weight of the pricing influencing factor, is the criterion layer comparison matrix, and .

[0071] Preferably, it is necessary to conduct a consistency test on the criterion layer comparison matrix . If the matrix meets the following conditions, it is a consistent matrix, which is the final weight of the pricing influencing factor, and the expression is as follows:

[0072] In actual situations, it is very difficult for the matrix to meet the consistency. Therefore, determine the allowable range of inconsistency through the consistency test. Among them, the consistency index expression is:

[0073] where is the consistency index, is the maximum eigenvalue of the criterion layer comparison matrix, is the matrix order; when When it is, it means complete consistency; when is close to 0, it is relatively consistent; when is larger, the inconsistency is more serious.

[0074] In order to quantitatively measure and the degree of inconsistency of the matrix, the random consistency index is introduced, which is obtained from random pairwise comparison matrices, and its formula is as follows:

[0075] Among them, is the number of random pairwise comparison matrices, is the order of the matrix, then the consistency ratio is:

[0076] Generally, when the consistency ratio is, it is considered that the degree of inconsistency of the matrix is within the allowable range, regarded as passing the consistency test, and can be used as the weight of the influencing factors. Otherwise, it is necessary to reconstruct the comparison matrix of the criterion layer and adjust in .

[0077] By dynamically adjusting the weights of multi-dimensional computing power resource pricing factors through the analytic hierarchy process, the computing power resource pricing can respond to market fluctuations and changes in user needs in real time, improving the scientificity and reliability of the computing power resource pricing results.

[0078] Preferably, as Figure 7 shown, the steps for estimating the computing power resource pricing in S400 include: Obtain the pre-estimated computing power resource pricing based on the pre-estimated unit cost of the computing power resource and the weights of the pricing influencing factors, and the expression is as follows:

[0079]

[0080] Among them, is the pre-estimated computing power resource pricing, is the profit, is the pre-estimated unit cost of the computing power resource, is the profit margin of the computing power market, is the parameter of the user configuration factor, is the weight of the pricing influencing factor.

[0081] Perform a difference check between the pre - estimated force resource pricing and the average historical computing power resource price, that is, use a comparator to determine whether the pre - estimated force resource pricing approaches the historical computing power resource price used at the first estimated running time. , and the difference check expression is as follows:

[0082] Wherein, is a preset threshold. When , the pre - estimated force resource pricing is reasonable; when , then replace the historical price average value with the pre - estimated force resource pricing , recalculate the unit cost of the pre - estimated force resource , and iterate until the difference meets the preset threshold to obtain the final computing power resource pricing.

[0083] In some embodiments, the preset threshold is set to 0.05. In typical scenarios, only a few iterations are required to converge the computing power resource pricing result to a reasonable range, significantly improving the computing efficiency and ensuring the rapid scheduling of the supercomputer center resources.

[0084] Embodiment 2: Figure 8 This is a schematic diagram of the computing power resource pricing system for the supercomputer internet in the embodiments of the present invention. As Figure 4 shown, the present invention provides a computing power resource pricing system for the supercomputer internet, including: a fixed cost accounting module 100 for computing power resources, a unit cost accounting module 200 for computing power resources, a weight acquisition module 300 for factors influencing computing power resource pricing, and a pricing estimation module 400 for computing power resources.

[0085] The fixed cost accounting module 100 for computing power resources: Obtain the variable cost per unit running time and the annual fixed cost based on the supercomputer center operation records, and convert the annual fixed cost to obtain the annual fixed cost depreciation amount; The unit cost accounting module 200 for computing power resources: Obtain the estimated annual running time through cubic spline interpolation based on the average historical computing power resource price and historical running time data; Obtain the unit cost of the pre - estimated computing power resources based on the variable cost per unit running time, the annual fixed cost depreciation amount, and the estimated annual running time; The weight acquisition module 300 for factors influencing computing power resource pricing: Construct a hierarchical structure model based on the analytic hierarchy process, and obtain the weights of the factors influencing pricing based on the hierarchical structure model; The pricing estimation module 400 for computing power resources: Obtain the pre - estimated computing power resource pricing based on the unit cost of the pre - estimated computing power resources and the weights of the factors influencing pricing; Perform a difference check between the pre-estimated computing power resource pricing and the average value of the historical computing power resource prices until the difference meets a preset threshold, and obtain the final computing power resource pricing. Embodiment 3: As shown in combination with Figure 9 This embodiment discloses a specific implementation manner of a computer device. The computer device may include a processor 41 and a memory 42 storing computer program instructions.

[0086] Specifically, the above-mentioned processor 41 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0087] Among them, the memory 42 may include a mass storage for data or instructions. By way of example and not limitation, the memory 42 may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disc, a magneto-optical disc, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. In suitable cases, the memory 42 may include removable or non-removable (or fixed) media. In suitable cases, the memory 42 may be internal or external to the data processing device. In a particular embodiment, the memory 42 is a non-volatile memory. In a particular embodiment, the memory 42 includes a read-only memory (ROM) and a random access memory (RAM). In suitable cases, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM), or a flash memory, or a combination of two or more of these. In suitable cases, the RAM may be a static random access memory (SRAM) or a dynamic random access memory (DRAM), where the DRAM may be a fast page mode dynamic random access memory (FPMDRAM), an extended date out dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0088] The memory 42 can be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 41.

[0089] The processor 41 reads and executes the computer program instructions stored in the memory 42 to implement the computing power resource pricing method for the supercomputer Internet in the above embodiments.

[0090] In some of the embodiments, the computer device may further include a communication interface 43 and a bus 40. Among them, as Figure 9 shown, the processor 41, the memory 42, the communication interface 43, and the bus 40 are connected and communicate with each other.

[0091] The communication interface 43 is used to implement communication between the various modules, devices, units, and / or devices in the embodiments of the present application.

[0092] The communication interface 43 can also implement data communication with other components such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations.

[0093] The bus 40 includes hardware, software, or both, and couples components of a computer device to each other. The bus 40 includes, but is not limited to, at least one of the following: Data Bus, Address Bus, Control Bus, Expansion Bus, and Local Bus. By way of example and not limitation, the bus 40 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. In suitable cases, the bus 40 may include one or more buses. Although the embodiments of the present application describe and illustrate specific buses, the present application contemplates any suitable bus or interconnect.

[0094] In addition, in combination with the computing power resource pricing method for the supercomputer Internet in the above embodiments, an embodiment of the present application can be implemented by providing a computer-readable storage medium. Computer program instructions are stored on the computer-readable storage medium; when the computer program instructions are executed by a processor, any one of the computing power resource pricing methods for the supercomputer Internet in the above embodiments is implemented. The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A computing power resource pricing method for supercomputing Internet, characterized in that: include: Computing power resource fixed cost accounting steps: obtain the variable cost per unit operating time and the annual fixed cost based on the supercomputing center operation record, and convert the annual fixed cost to obtain the annual fixed cost depreciation amount; Unit cost calculation steps for computing power resources: Based on the historical average computing power resource price and historical running time data, the estimated annual running time is obtained through cubic spline interpolation method; Obtaining an estimated unit cost of human resources based on the variable cost per unit operating time, the annual fixed cost depreciation amount, and the estimated annual operating time; The step of obtaining the weights of the factors influencing the pricing of computing power resources is as follows: constructing a hierarchical structure model based on the hierarchical analysis method, and obtaining the weights of the factors influencing the pricing based on the hierarchical structure model; The computing power resource pricing estimation step: obtaining the estimated computing power resource pricing based on the estimated computing power resource unit cost and the pricing influencing factor weights; The estimated computing resource pricing is checked against the average historical computing resource price until the difference meets a preset threshold, and the final computing resource pricing is obtained.

2. The method for pricing computing resources for supercomputing Internet according to claim 1 is characterized in that: The computing power resource unit cost calculation step includes: definition The price of computing resources and running time The cubic spline interpolation function between Upper Division Sample points ; The cubic spline interpolation function The following conditions must be met: The interpolation conditions are: ; The continuity condition is: at the endpoints of the interval Department, Continuous and first-order derivative , second-order derivative continuous; The boundary conditions are: and ; Based on the interpolation condition, the continuity condition and the boundary condition, a cubic spline equation group is established, and a cubic spline interpolation function is obtained by solving the equation group. ; Based on the historical price average Pp, obtain the estimated annual operating time .

3. The method for pricing computing resources for supercomputing Internet according to claim 2 is characterized in that: The computing power resource unit cost accounting step further includes: Based on the annual fixed cost depreciation amount and the estimated annual operating time, the fixed cost per unit operating time is calculated as follows: in, is the fixed cost per unit running time, is the annual fixed cost depreciation amount, It is the estimated annual operating time; Based on the fixed cost per unit running time and the variable cost per unit running time, the unit cost of computing power resources is calculated, and the expression is as follows: in, is the unit cost of computing resources, is the fixed cost per unit running time, is the variable cost per unit of operating time.

4. The method for pricing computing resources for supercomputing Internet according to claim 3 is characterized in that: The step of obtaining the weights of the factors affecting the pricing of computing power resources includes: Establishing a hierarchical model step: constructing the hierarchical model, wherein the hierarchical model includes a computing power pricing target layer and a criterion layer connected in sequence, wherein the criterion layer has multiple pricing influencing factors, including: scenario factors, chip factors, storage factors, network factors, computing factors, and equipment factors; Constructing a comparison matrix step: comparing multiple pricing influencing factors in the criterion layer in pairs based on a 1-9 scaling method to construct a criterion layer comparison matrix; Weight calculation step: Calculate the weight of pricing influencing factors through the criterion layer comparison matrix based on the geometric mean method, expressed as: in, is the weight of pricing factors, is the criterion level comparison matrix, and .

5. The method for pricing computing resources for supercomputing Internet according to claim 4 is characterized in that: The step of obtaining the weights of the factors affecting the pricing of computing power resources further includes: Performing a consistency check on the criterion layer comparison matrix; Consistency Index for: in, is the consistency indicator, is the maximum eigenvalue of the criterion layer comparison matrix, is the matrix order; random consistency index for: in, is the number of random criterion layer comparison matrices, is the matrix order; Consistency ratio for: When the consistency ratio , the consistency check is passed, otherwise, the criterion layer comparison matrix is ​​reconstructed.

6. The method for pricing computing resources for supercomputing Internet according to claim 5 is characterized in that: The computing power resource pricing estimation step includes: The estimated labor resource pricing is obtained based on the estimated unit cost of the labor resource and the weight of the pricing influencing factors, and the expression is as follows: in, pricing the estimated human resources, For profit, is the estimated unit cost of human resources, is the computing power market profit margin, Configure factor parameters for users, is the weight of the pricing influencing factor.

7. The method for pricing computing resources for supercomputing Internet according to claim 6 is characterized in that: The computing power resource pricing estimation step further includes: The estimated computing resource pricing is checked against the historical computing resource price average, and the difference check expression is as follows: in, is the preset threshold, when When , the estimated human resource pricing is reasonable; when When the estimated human resource pricing Replace the historical price average , recalculate the estimated unit cost of human resources , iterate until the difference meets the preset threshold, and obtain the final computing power resource pricing.

8. A computing resource pricing system for supercomputing Internet, characterized in that: The computing power resource pricing method for supercomputing Internet as applied to any one of claims 1 to 7 above comprises: Computing resource fixed cost accounting module: obtains the variable cost per unit operating time and the annual fixed cost based on the operation records of the supercomputing center, and converts the annual fixed cost to obtain the annual fixed cost depreciation amount; Computing power resource unit cost accounting module: Based on the historical computing power resource price average and historical running time data, the estimated annual running time is obtained through the cubic spline interpolation method; Obtaining an estimated unit cost of human resources based on the variable cost per unit operating time, the annual fixed cost depreciation amount, and the estimated annual operating time; The computing power resource pricing influencing factor weight acquisition module: constructs a hierarchical structure model based on the hierarchical analysis method, and obtains the pricing influencing factor weights based on the hierarchical structure model; A computing power resource pricing estimation module: obtaining an estimated computing power resource pricing based on the estimated computing power resource unit cost and the weight of the pricing influencing factors; The estimated computing resource pricing is checked against the average historical computing resource price until the difference meets a preset threshold, and the final computing resource pricing is obtained.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it implements the computing power resource pricing method for supercomputing Internet as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, a computing power resource pricing method for supercomputing Internet as described in any one of claims 1 to 7 is implemented.

Images