Information processing device, information processing method, and program

Abstract

To improve the convenience of providing, using, and managing computing resources. The server 1 manages the buying and selling of computing power between a computing power user U who uses computing power to execute a computing job and a computing power owner O who provides computing power, and includes a price setting and order management unit 41. The price setting and order management unit 41 evaluates a predetermined value (e.g., XX points, etc.) for a computing power ticket, which is a right to use a predetermined unit of computing power defined by multiple indicators (e.g., power, hardware, software, data) provided by the computing power owner O.

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] Conventionally, when jobs submitted by a plurality of users are executed using a plurality of servers, there is a technique for managing restrictions on the number of server computers assigned to each user (for example, see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2007-179100 A Summary of the Invention [Problem to be solved by the invention]

[0004] Users are not always in a situation where they need to process a certain amount of jobs. To address this issue, there is a method of reducing costs by making a contract to use servers on a cloud service and changing the number of servers available. In addition, it is time-consuming for providers of such cloud services to secure and manage appropriate computing resources.

[0005] The present invention has been made in consideration of the above circumstances, and has an object to improve the convenience of providing, using, and managing computational resources. [Means for solving the problem]

[0006] In order to achieve the above object, an information processing system according to one aspect of the present invention comprises: a computing power value evaluation means for evaluating a predetermined value for a right to use a predetermined unit of computing power defined by a plurality of indexes provided by a computing power owner; Equipped with.

[0007] An information processing method and a program according to one aspect of the present invention are a method and a program corresponding to the information processing device according to the aspect of the present invention described above. Effect of the Invention

[0008] According to the present invention, it is possible to improve the convenience of providing, using, and managing computational resources. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram showing an overview of a present service to which an information processing system including a server according to an embodiment of an information processing device of the present invention is applied. [Diagram 2] 2 is a diagram illustrating an example of the configuration of an information processing system that realizes the present service illustrated in FIG. 1. [Diagram 3] 3 is a block diagram showing an example of a hardware configuration of a server in the information processing system of FIG. 2. [Figure 4] 4 is a functional block diagram showing a first embodiment of the functional configuration of a server having the hardware configuration of FIG. 3. [Diagram 5] 4 is a diagram for explaining the flow of processing of computational power trading in a server having the functional configuration of FIG. 3. [Figure 6] 6 is an arrow chart showing in detail the flow of the computational power transaction in FIG. 5 . [Figure 7] FIG. 2 is a diagram showing the configuration of a computing power ticket used in the present service shown in FIG. 1. [Figure 8] 8 is a diagram showing an example of a calculation power coupon used in the components of the calculation power coupon of FIG. 7. [Figure 9] 8 is a diagram showing another example of a calculation power ticket used in the components of the calculation power ticket of FIG. 7. FIG. [Figure 10] This is a diagram for explaining the transition of components of computational power in a computational power ticket transaction, divided into phases. [Figure 11] 4 is a functional block diagram showing a second embodiment of the functional configuration of a server having the hardware configuration of FIG. 3. [Figure 12]12 is a diagram for explaining the flow of processing of computational power trading in a server having the functional configuration of FIG. 11. [Figure 13] FIG. 3 is a diagram showing an example of a simple flow of computational power trading in the information processing system shown in FIG. 2. [Figure 14] FIG. 3 is a diagram showing an example of computing power trading in insurance services in the information processing system shown in FIG. 2. [Figure 15] FIG. 13 is a diagram showing an example in which a rating function is added to evaluate the computational resources provided by a computational power owner O. [Figure 16] FIG. 16 is a diagram showing an example in which an investor element is further added to the example in FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. First, referring to FIG. 1, an overview of a service (hereinafter referred to as "this service") that can be realized by an information processing system (see FIG. 2 described later) to which a server according to one embodiment of the present invention is applied will be described.

[0011] FIG. 1 is a diagram showing an overview of a service (hereinafter, referred to as "this service") that can be realized by an information processing system to which a server according to an embodiment of an information processing device of the present invention is applied.

[0012] As shown in Figure 1, this service is a service in which a server 1 managed by a service provider S mediates (matches) the buying and selling of computing power between a computing power owner O who provides computing power and a computing power user U who needs computing power.

[0013] The server 1 accepts the provision of computing power from the computing power owner O, and provides it to the computing power user U (receiving computing power from the perspective of the computing power user U). The server 1 also accepts funds from the computing power user U for obtaining the computing power that the computing power user U desires, and pays to the computing power owner O a price according to the computing power traded between the computing power user U and the computing power owner O.

[0014] Here, calculation refers to numerical calculations that have industrial or academic value, such as machine learning learning and inference, simulation, statistical processing, visualization processing, etc. Computational power refers to elements such as goods and information that are essential or desirable to use when performing calculations. Specific elements that make up computational power are, for example, parts or combinations of power, hardware (HW), software (SW), data, etc. Details of the components of computational power will be described later.

[0015] A computing power owner O is a person who owns at least a portion of computing power. A single computing power owner O may own all of the components of computing power that are sufficient to perform one computation, or may own some of the components of computing power such that a computation can be performed by combining the computing power owned by multiple computing power owners O.

[0016] A computing power user U may perform calculations by receiving components of computing power provided by multiple computing power owners O, or may perform calculations by receiving computing power provided by a single computing power owner O. Also, a computing power user U may perform calculations by obtaining a part of the computing power from the computing power owner O and combining it with a part of the computing power that the computing power user U has.

[0017] The flow of the service will be specifically explained below with reference to FIG. In step S1, a computing power owner O provides information of computing power to a server 1. In step S2, the server 1 receives information on the computing power provided by the computing power owner O. In step S3, the computational power user U inputs funds and conditions for obtaining computational power to the server 1. In step S4, the server 1 manages funds input by the computing user U.

[0018] Here, in step S5, the server 1 evaluates the computing power provided by the computing power owner O in terms of a predetermined value, such as points, for the right to use a predetermined unit of computing power defined by multiple indicators (e.g., power, hardware, software, data). In step S6, the server 1 issues a computing power ticket to the computing power owner O for the computing power owner O to exercise the right in accordance with the evaluated value.

[0019] In step S7, the server 1 carries out buying and selling transactions of computing power. Specifically, the server 1 executes control to put up a computational power ticket provided by a computational power owner O and to allow a computational power user U among a plurality of computational power users U who satisfies a predetermined bidding condition to win the computational power ticket. When buying and selling computational power, the server 1 selects a computational power ticket that is suitable for the conditions input by the computational power user U, and recommends the computational power ticket to the computational power user U. During a buying and selling transaction, when multiple computing power tickets (with the same specifications) with at least some of the same indicators are available for bidding, the server 1 performs control to allow the computing power user U to bid for (receive) a specified unit of computing power by giving priority to a computing power ticket with a high predetermined priority (for example, a computing power ticket of a computing power owner O whose resource was registered early) among the multiple computing power tickets.

[0020] In step S8, when the computing power that meets the conditions of the computing power user U is sold, the server 1 evaluates the value of the sold computing power ticket based on the winning conditions. This evaluation can be used as an index for assigning a value the next time the same computing power is offered. Furthermore, the server 1 executes control to transfer the computing power corresponding to the computing power ticket from the computing power owner O to the computing power user U. The server 1 executes control to make the computing power user U to whom the predetermined unit of computing power has been transferred receive the predetermined unit of computing power indicated on the computing power ticket.

[0021] As a result, in step S9, the computing power is provided from the computing power owner O to the computing power user U. From the perspective of the computing power user U, upon receiving the provision of the computing power, as in step S10, the computing power user U receives the computing power and can thereafter use the computing power.

[0022] In step S11, the server 1 executes a process of paying the computational power owner O a compensation (for example, points) for the computational power provided to the computational power user U from the funds input by the computational power user U.

[0023] As a result, in step S12, the computing power owner O can receive compensation for the computing power provided to the computing power user U.

[0024] In this way, according to this service, the server 1 mediates (matches) the buying and selling of computing power between a computing power owner O who provides computing power and a computing power user U who needs computing power, evaluates the value of the computing power provided by the computing power owner O, issues a computing power ticket of that value, and allows the computing power user U to conduct buying and selling transactions in units of the computing power ticket, thereby making it possible to improve the convenience of the provision, use, and management of computing resources.

[0025] Next, with reference to FIG. 2, a description will be given of the configuration of an information processing system that realizes the provision of the above-mentioned service, that is, an information processing system including a server according to an embodiment of an information processing device of the present invention. FIG. 2 is a diagram showing an example of the configuration of an information processing system including a server according to an embodiment of the information processing device of the present invention.

[0026] The information processing system shown in FIG. 2 is configured to include a server 1, a user terminal 2, and an owner terminal 3. The server 1, the user terminal 2, and the owner terminal 3 are connected to each other via a predetermined network NW such as the Internet.

[0027] The server 1 is an information processing device managed by a service provider. The server 1 executes various processes for realizing this service while appropriately communicating with the user terminal 2 and the owner terminal 3.

[0028] The user terminal 2 is an information processing device operated by a computing user U. The user terminal 2 is configured with a smartphone, a tablet, a personal computer, or the like. The owner terminal 3 is an information processing device operated by a computing owner O. The owner terminal 3 is configured with a smartphone, a tablet, a personal computer, or the like.

[0029] FIG. 3 is a block diagram showing an example of a hardware configuration of a server in the information processing system shown in FIG.

[0030] The server 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a bus 14, an input / output interface 15, an input unit 16, an output unit 17, a memory unit 18, a communication unit 19, and a drive 20.

[0031] The CPU 11 executes various processes according to a program recorded in the ROM 12 or a program loaded from the storage unit 18 into the RAM 13 . The RAM 13 also stores data and the like necessary for the CPU 11 to execute various processes.

[0032] The CPU 11, ROM 12, and RAM 13 are connected to one another via a bus 14. An input / output interface 15 is also connected to this bus 14. An input unit 16, an output unit 17, a storage unit 18, a communication unit 19, and a drive 20 are connected to the input / output interface 15.

[0033] The input unit 16 is configured with, for example, a keyboard and is used to input various information. The output unit 17 is configured with a display such as a liquid crystal display, a speaker, and the like, and outputs various information as images and sounds. The storage unit 18 is configured with a dynamic random access memory (DRAM) or the like, and stores various data. The communication unit 19 communicates with other devices (for example, the user terminal 2 and the owner terminal 3) via a network NW including the Internet.

[0034] Removable media 21, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is appropriately loaded into the drive 20. A program read from the removable media 21 by the drive 20 is installed in the storage unit 18 as necessary. Further, the removable medium 21 can store various data stored in the storage unit 18 in the same manner as the storage unit 18 .

[0035] Although not shown, the user terminal 2 and the owner terminal 3 shown in Fig. 2 can have a configuration basically similar to the hardware configuration shown in Fig. 3. Therefore, a description of the hardware configuration of the user terminal 2 and the owner terminal 3 will be omitted.

[0036] 2, including the server 1 as shown in Fig. 3, various hardware and software components work together to execute various processes for buying and selling computing power between the user terminal 2 and the owner terminal 3. As a result, the service provider S can provide the above-mentioned service to the computing power user U and the computing power owner O.

[0037] FIG. 4 is a functional block diagram showing an example of the functional configuration of the server in FIG. 2 and FIG. As shown in FIG. 4, the storage unit 18 of the server 1 stores a database 90 (hereinafter referred to as "DB 90"). DB90 stores information regarding computational power transactions, including personal information provided or input by each of the computational power owner O and computational power user U (such as name, input funds, computational power tickets, points held, available computational power, etc., associated with the account (ID) used for the transaction). The DB 90 stores a predetermined value, such as points, evaluated for the right to use a predetermined unit of computing power defined in the "information on computing power" provided by the computing power owner O. "Information about computing power" is a number of indicators that make up computing power (for example, power, hardware, software, data, etc.). The DB 90 stores computing power coupons issued to the computing power owner O in order for the computing power owner O to exercise his / her rights according to the evaluated points. The computing power coupons are used in buying and selling transactions of computing power. The DB 90 stores information indicating the correspondence between the calculation power ticket and the points evaluated therefor. DB90 stores priority conditions for executing control such that when multiple calculation power coupons of the same index (e.g., the same specifications) are traded (used) in a transaction, a calculation power coupon with a higher predetermined priority among the multiple calculation power coupons is given priority and a calculation power user U receives a predetermined unit of calculation power. The condition for a computing power ticket with a high predetermined priority is, for example, a computing power ticket for which the computing power owner O registered resources early, etc. Alternatively, other conditions may be used.

[0038] As shown in FIG. 4, in the CPU 11 of the server 1, a selling price management unit 30 and a computing power management unit 50 function when executing processing related to buying and selling transactions of computing power.

[0039] The selling price management unit 30 includes a price setting and order management unit 41 and a fund management unit 43 . The price setting and order management unit 41 evaluates the value of a computational power ticket that utilizes a predetermined unit of computational power (e.g., computational power that can be provided by the computational power owner O) defined by multiple elements (indicators) provided by the computational power owner O, and then performs a process of buying and selling the computational power ticket at that value.

[0040] In addition, the price setting and order management unit 41 evaluates the value of the successful bid computing power ticket based on predetermined bidding conditions. As a result, the value at the time of evaluation is reflected in the evaluation of computing power newly provided by the computing power owner O in the future. As a result, if the successful bid computing power ticket is highly evaluated, the evaluation of the computing power owner O improves, and the value (price) of the computing power provided by the computing power owner O can be increased. The price setting and order management unit 41 may use tokens, such as points, as a unit of transaction.

[0041] The price setting and order management unit 41 selects a computational power coupon that meets the conditions specified by the computational power user U, and recommends the computational power coupon to the computational power user U. This allows a calculation power user U to easily find a calculation power ticket containing a desired calculation power among the many calculation power tickets on offer, without having to check the contents of each ticket, which vary slightly in calculation power. As a recommendation function, search software, simulation software, etc. may be installed in advance on the server 1, and the software may be operated in response to instructions from the pricing and order management unit 41 to select candidate calculation results from the DB 90.

[0042] The fund management unit 43 executes a process of paying the price of the computational power ticket that the computational power user U has won to the computational power owner O. Specifically, the fund management unit 43 manages, in each account, the funds input by the computing power user U, the price (consideration) paid for purchasing computing power from the funds, the remaining amount after deducting the price from the funds, and the amount acquired by the computing power owner O who received the price.

[0043] The computing power management unit 50 provides computing power to a computing power user U that meets the conditions (items written on a computing power ticket) specified by a computing power owner O. When a computational power ticket having at least some of the same computational power elements (indicators) is used (by winning a bid), the computational power management unit 50 executes control to give priority to a computational power owner O of a predetermined computational power ticket among the multiple computational power tickets, and to have the computational power user U receive the computational power ticket (a predetermined unit of computational power) won by the computational power user U. A computational power ticket with a high predetermined priority is, for example, a computational power ticket of a computational power owner O whose resource registration is early. Note that priority may be given not only to the computational power owner O, but also to the computational power user U. The computational power management unit 50 may use a token, such as points, as a transaction unit.

[0044] The computational power management unit 50 includes a computational power providing management unit 61 and a computational power receiving management unit 62 . The computing power provision management unit 61 manages the computing power provided by the computing power owner O. Specifically, the computing power provision management unit 61 receives the content of the provision of computing power by the computing power owner O, "information on computing power." The received content is stored in the DB 90 and is shared so as to be available to other units.

[0045] The computing power receipt management unit 62 provides the computing power user U, via the computing power owner O, with computing power equivalent to the computing power ticket that the computing power user U has made a successful bid for.

[0046] Thus, according to the configuration of the server 1 of the first embodiment, the server 1 issues a computing power ticket with a value obtained by evaluating a predetermined unit of computing power defined by a plurality of indicators (e.g., power, hardware, software, data, etc.) provided by the computing power owner O according to a certain standard, and by buying and selling the computing power ticket, the computing power owner O can freely provide as much computing power as he or she can provide. Furthermore, by buying and selling the computing power ticket, the computing power user U can purchase and use computing power of the desired quality and quantity. As a result, it is possible to improve the convenience of both the computing power owner O and the computing power user U with regard to the utilization of computing power, including the provision, utilization, and management of computing resources.

[0047] Next, the operation of the server of the first embodiment will be described with reference to FIG. 5 and FIG. FIG. 5 is a diagram for explaining the flow of a calculation power transaction in a server having the functional configuration (first embodiment) of FIG. FIG. 6 is an arrow chart showing in detail the flow of the computational power transaction in FIG.

[0048] In the case of the server of the first embodiment, in step S21 shown in FIG. 5, when the owner terminal 3 provides the server 1 with computing power through the operation of the computing power owner O, the computing power provision management unit 61 in the server 1 accepts the computing power. Providing computing power may mean making the components of computing power available for use. It may also mean making the components of computing power available for use by a computing power user U at a future date and time or for a future period of time by contract or the like. Making the components of computing power available to a computing power user U can include, for example, entering into a power grid transmission and reception contract, providing power generation equipment such as power generation facilities and power storage facilities, providing a server, providing server components such as CPUs and GPUs, providing means for using the server such as issuing SSH keys, providing software, issuing software licenses, providing data, and providing security measures such as locking and unlocking related items. The computing power provision management unit 61 accepts the above-mentioned provision contents from the computing power owner O. The accepted contents are stored in the DB 90 and shared with the selling price management unit 30. Here, sharing means that the selling price management unit 30 is notified by the computing power provision management unit 61 that computing power information has been provided by the computing power owner O, and the contents are made available for reading from the DB 90 by the selling price management unit 30 at any time. At this time, the computing power owner O may accept sales conditions such as a desired selling price of the computing power. The selling price management unit 30 issues a computing power ticket indicating the shared content of the computing power provision. The computing power ticket is preferably electronic. Specifically, the issuing information of the computing power ticket is shared with the price setting and order management unit 41. The computing power ticket includes a desired selling price by the computing power owner O.

[0049] In step S22, the computing power user U operates the user terminal 2 to input funds for purchasing computing power to the server 1 by transmitting the funds from the user terminal 2. In the server 1, the fund management unit 43 accepts the funds transmitted from the user terminal 2. At this time, the fund management unit 43 accepts the contract for sending or receiving cash or money and the amount. The accepted information such as the amount is stored in DB90, and the storage location of the information is notified to the price setting and order management unit 41 of the selling price management unit 30 and shared. When investing funds, the computing power user U can specify the purchase conditions for the computing power (for example, minimum conditions and desired conditions for each component, desired date and time and period for using the computing power, etc.) as necessary. This information is also stored in the DB 90 and shared with the selling price management unit 30.

[0050] In step S23, in selling price management unit 30, price setting and order management unit 41 determines the selling price by a predetermined method based on the shared calculation information (contents), and matches it with the amount received separately. The selling price may be determined, for example, by pre-determining an absolute fixed price based on a quantitative and qualitative evaluation of the computing power provided, or it may be determined relatively by comparing it with other computing powers accepted at a given time. Furthermore, when matching, it is not necessary that the only matching requirement for the components of computational power is monetary amount, and matching may be carried out by setting constraints such as minimum conditions and desired conditions for each component of computational power. In this case, for example, the pricing and order management unit 41 can perform a matching process that takes into account constraints by disclosing the details of the provided computational power to the computational power user U and accepting the quantitative and qualitative evaluation of the provided computational power by the computational power user U. As a result of the above, if the computing power and the amount match, the sale of computing power is completed.

[0051] In step S24, the sale of computing power is concluded, that is, the computing power is sold, and the matching details are shared from the price setting and order management unit 41 to the computing power receipt management unit 62.

[0052] In step S25, the computing power receiving management unit 62 notifies the user terminal 2 that the computing power has been sold, and the computing power user U can receive the computing power provided by the computing power receiving management unit 62. This allows the computing power user U to perform calculations using the received computing power. In addition, if necessary, computing power may be provided directly from the computing power owner O, in which case the computing power provision management unit 61 and the computing power receipt management unit 62 are notified that computing power will be provided directly from the computing power owner O to the computing power user U. In addition, if a bidder wins a bid for computing power that can be used at a specified date and time in the future, the computing power will be provided at that date and time when it can be used. In other words, the computational power that can be used in a specified future period is provided at an appropriate time based on the policy set by the computational power owner O when providing the computational power and the date, time, and period desired by the computational power user U.

[0053] In step S26, the fund management unit 43 pays the computational power owner O a fee based on the content of the matching that has been established. This allows the computing power owner O to receive the payment from the fund management unit 43.

[0054] The above describes the flow of the present service by the server 1, taking the procedure of steps S21 to S26 as an example. Specifically, the server 1 can provide the present service through a series of steps, such as provision of computing power from the computing power owner O, receipt of computing power from the computing power user U, acceptance of the consideration from the computing power user U, and payment of the consideration to the computing power owner O.

[0055] Here, the process of determining the selling price (sales price) of a computing power ticket in the flow of computing power trading will be described with reference to FIG.

[0056] First, in step PO1, the server 1 registers in the DB 90 the information on the computing power provided by the computing power owner O (contents of the computing power), the terms of sale, and the selling price. The selling price may be set as a fixed price like a limit order, or as a market order, set to any price or higher (for example, 100 points or more, 0 points or more, etc.), and the transaction may be set to proceed as soon as the buyer's calculation power user U is generated. In addition, the calculation power ticket also includes the calculation power execution conditions (including the period, start time, etc.). When the buyer is choosing the computing power and user U's environment due to physical or legal constraints (such as line problems or export control issues), it is acceptable to set purchase conditions for the buyer so that the transaction is not necessarily concluded based on price alone.

[0057] In step PC1, the above-mentioned registration process enables issuance of a computing power ticket. The specific content of the computational power is written on the computational power ticket, and a selling price is set. When computational power tickets are collected from various computational power owners O, in step PC2, the multiple computational power tickets are displayed in the server 1 (a process of sorting the computational power tickets in order according to a certain index (order of reception, order of price, etc.)). This information can be observed by the computational power owner O. If necessary, in step PO2, the calculation power ticket can be observed and the buying and selling conditions can be updated. This allows the user to refer to many calculation power tickets in the market, and to withdraw calculation power tickets for the calculation power owner O who has displayed them, change the contents of the calculation power tickets (withdraw and re-register), revise the selling price, etc.

[0058] A computational ability user U observes the computational ability ticket at any time in step PU1, and examines the contents of the display of the computational ability ticket in step PC3. If the calculation power user U examines the offer and finds that there is no calculation power ticket that matches his / her desires, he / she gives up on the purchase in step PU2 and ends the transaction. On the other hand, if there is a calculation power ticket that matches the desired content, in step PU3, the purchase decision of the calculation power ticket and the order conditions are presented, and in step PC4, the bidding conditions of the calculation power ticket are matched. In matching the bidding conditions of the calculation power ticket, the price is basically presented, and the decision of whether to accept the bid is made mechanically. However, if there are purchase conditions for the buyer, a response to the details is received from the computing power user U in the process of step PU3, and the server 1 or the computing power owner O determines whether or not the bid is successful. If the result of the determination is that the bid is not successful, the transaction is terminated in step PU4.

[0059] Then, in step PC5, the successful bid is confirmed by the successful bid and the display is updated, and the successful bid is deleted from the display. At the same time, in step PU5, the successful bid is notified to the calculation power user U and the calculation power owner O in the successful bid notification step PO3. At the same time, a payment is made between the computing power user U and the server 1 in step PU6, and a payment is made between the server 1 and the computing power owner O in step PO4.

[0060] In step PU7, the user U of the computing power declares receipt of the computing power based on the implementation conditions of the computing power written on the computing power ticket. At this time, for items that can be systematically started by server 1 (for example, provision of a VM environment, provision of data, provision of electricity, lines, and cloud environment after appropriate configuration), server 1 provides computing power in step PC6. For transactions that cannot be started systematically on the server 1 (for example, rental of physical items, provision of a cloud environment where the computing power owner O manages whether or not access is possible, physical locking and unlocking operations, etc.), computing power is provided through the process of (notification of receipt declaration) from the server 1 to the computing power owner O in step PC7, (provision of computing power) from the computing power owner O to the computing power user U in step P05, and (provision notification) from the computing power owner O to the server 1 in step P06.

[0061] After the above, in step PU8, the computational power user U receives computational power, and in step PU9, the computational power is used. After use, in step PU10, the user U of computing power returns the computing power. Regarding the return, for those that can be systematically terminated by the server 1 (for example, cloud environments or licensed software that expire based on a date and time, etc.), in step PC8, the server 1 notifies the user U of the return / return of the computing power, so that the user U of computing power notifies or collects the computing power to the server 1.

[0062] When the server 1 needs to return the computing power to the computing power owner O (for example, to recover the license of the licensed software), the server 1 returns / notifies the computing power owner O of the return in step PC9.

[0063] After the notification, in step P07, the computing power owner O receives the computing power, and the process is completed. At this time, with regard to the things consumed by the computing power user U, such as electricity, only notification is performed. On the other hand, if there is something that needs to be returned directly from the computing power user U to the computing power owner O because the transaction cannot be completed systematically (for example, rental of an item), in step PC10, the computing power is returned from the computing power user U to the computing power owner O by returning the computing power. After that, the returned computing power can be provided again from the computing power owner O to the server 1. Through the above procedure, computing power is traded.

[0064] Here, the price determination process in the computational power trading shown in FIG. 6 will be described. In the case of the auction format (parts marked with * in FIG. 6), the computing power owner O can register desired conditions before the auction (minimum desired price, auction period, etc.) when providing computing power and presenting the buying and selling conditions in step PO1, or when observing the computing power ticket and updating the buying and selling conditions in step PO2.

[0065] In step PC4, in matching the bidding conditions for the computational power ticket, the server 1 accepts bids (presentations of order conditions) from computational power users U during a predetermined period. That is, in step PU3, the desired conditions of the calculation power user U are sent to the server 1 in deciding to purchase a calculation power ticket and presenting the order conditions. During a predetermined period, as in a general auction, desired conditions are accepted from multiple computational power users U, the current highest bid is disclosed, and multiple desired conditions are accepted from the same computational power user U, thereby widely accepting desired conditions. At the end of the period, the highest bidder is awarded the bid, and in step PC5, the bid is established and the display is updated. The other users U of computational ability end the transaction as the bidder is not successful in step PU4. Also, if no users U of computational ability show up, the transaction ends.

[0066] In an auction of computing power, the typical price for using computing power in each industry, or legal regulations regarding line communication speeds and data, may differ depending on the attributes of the computing power user U who may be the successful bidder (commercial user, academic user, public user, etc.) and geographical environment. In this case, the computational power owner O can limit the attributes and environment of the computational power user U and set the desired price in stages in steps PO1 and PO2. Such settings may be set individually as conditions at the time of the auction, or can be realized by setting in steps PO1 and PO2 that additional fees or additional work will be incurred as necessary after the successful bid.

[0067] This setting method allows smooth computational power trading while responding to the attributes and environments of a wide variety of computational power users U, and also allows a wide range of computational power users U and computational power owners O to be accepted. In addition, by determining the price in an auction format, it is possible to determine an appropriate price for computational power based on the supply and demand balance.

[0068] Next, components of the calculation power supply will be described with reference to FIGS. FIG. 7 is a diagram illustrating an example of components of a calculation power sheet. 8 and 9 are diagrams showing examples of calculation power tickets on which the calculation power of a desired component among the components in FIG. 7 is written. The components of the calculation power card are individually defined items, information, etc. that are essential or desirable to use when performing a calculation.

[0069] Specifically, the components of a computing power ticket can be classified and defined as, for example, power, hardware (HW), software (SW), data, region, user type, ticket purchase timing, scheduled time period for calculation execution, reliability, etc., as shown in Figure 7. The computational power ticket functions as a promissory note indicating the content of the computational power provided by the computational power owner O, and a provision contract is established by buying and selling the computational power ticket.

[0070] A computing power owner O can convert the numerous computing resources he owns into computing power tickets in any unit and quantity. For example, one HW32 core server node may be configured as one HW32 core computing power ticket, or it may be configured as eight HW4 core computing power tickets assuming use in a VM (virtual machine) or container.

[0071] The computing power ticket may be created for each category based on items and information of a single category, or may be created by combining items and information of multiple categories, depending on the content provided by the computing power owner O. For example, any one server node may be configured as one HW computing power ticket, or if the computing power owner O sets the OS and drivers in advance and restricts the use thereon, one HW+SW computing power ticket may be configured.

[0072] Taking a single-category computational power voucher as an example, the computational power voucher can describe specific characteristics of the components of the computational power voucher so that the computational power user U can refer to the computational power voucher when selecting the computational power voucher. Below are examples of contents that can be described on a computational power voucher, including power, hardware (HW), software (SW), data, region, user type, ticket purchase timing, scheduled time period for computation execution, and reliability.

[0073] ·Electricity Able to describe power generation methods. For example, general electricity purchased from a power sharing business operator, self-generated electricity generated by self-owned facilities such as solar power generation, green electricity from wind, solar, wave power, etc. Able to describe fee structures. Price changes and dynamic pricing according to use, capacity, time, etc. Able to describe supply stability. Forecasts of power generation based on hours of sunshine in the case of solar power, forecasts of power shortages in local general electricity jurisdictions, etc.

[0074] Hardware (HW) Be able to describe machines, such as the specifications, quantity, and usage history of CPU, GPU, memory, etc., and the specifications of storage devices such as HDD and SSD. Be able to describe Internet lines, such as the specifications of installed lines such as LAN, contracted providers, and whether or not a VPN is used.

[0075] · Software (SW) Able to describe the OS. This is the software used by the user, which could be, for example, Linux or Windows. Note that Linux and Windows are registered trademarks. Able to describe drivers. For example, drivers that allow the use of some kind of HW or SW. Able to describe analysis software. For example, it could be simulation software that runs on the OS, or machine learning software. Able to describe middleware.

[0076] ·data Able to describe design data. For example, data such as drawings used in simulations, and various data such as images and audio used in machine learning. Able to describe analysis models. These can be models that define various physical phenomena in simulations, or models such as deep learning used in machine learning. Able to describe analysis conditions. For example, various conditions when running simulations or machine learning.

[0077] Ticket purchase timing The timing of ticket purchase can be described. Dynamic pricing may be based on the timing of the scheduled calculation execution date, such as 3 months, 1 month, 1 week, or the day itself. In addition, a cancellation policy may be set, and the price offered may be changed depending on whether the purchase of the calculation ticket can be canceled after it is purchased and before the calculation is executed. This cancellation may be related to cancellation due to the convenience of the calculation power user U or cancellation due to the convenience of the calculation power owner O.

[0078] Scheduled execution time You can specify the time period for which the calculation will be performed. You can change the price according to demand or set dynamic pricing during the daytime (9:00-17:00) on weekdays or holidays, at night (17:00-22:00), late night (22:00-5:00 the next morning), or early morning (5:00-9:00).

[0079] Reliability Be able to write about security. Specifically, the physical security of the server itself, the security of the software and virtual environment, and the security of communications. Also be able to write about ratings. Examples of ratings include the computing power of the server itself, the speed of communications, and the robustness of security. Able to record past performance. Also able to record availability such as the percentage of usage in the past and the calculation and communication speeds achieved in the past. Availability is, for example, the time a server has been operating without power outages, or the percentage of time it has been operating under appropriate temperature and humidity conditions. Able to record owner information. For example, whether it is owned by an individual, a corporation, a research institute, a university, etc.

[0080] User type You can specify the target users, such as corporations, academics, individuals, etc. You can set the price according to the type of user, such as offering a lower price for academic users than the general price.

[0081] Region You can specify the location of the server. For example, Japan (Tokyo, Osaka, etc.), America, Singapore, China, etc.

[0082] Here, an example of the calculation power ticket will be described with reference to FIG. 8 and FIG. The computing power ticket 91 in the example of FIG. 8 lists the machine with the specifications desired by the user, the power required to operate the machine, and the timing of ticket purchase. The computing power ticket 92 in the example of FIG. 9 lists the items in which a VM is installed on a machine, the OS, analysis software, and analysis data are stored in the VM, and the region of the items. By providing the computing power of the computing power tickets 91 and 92 as shown in the examples of Figures 8 and 9, the computing power user U can easily use the computing power. In addition, by recording the above-mentioned contents on the computing power ticket, the transaction between the computing power owner O and the computing power user U can be smoothly carried out. However, even if the same HW and power as the computing power ticket 91 in the example of FIG. 8 is specified, the offered price may be different depending on whether the ticket is purchased one month in advance or one week in advance. Similarly, even if the same VM, analysis data, analysis software, and OS as those in the computing power ticket 92 in the example of FIG. 9 are specified, the amount will vary if the region is changed.

[0083] The price of this calculation power ticket is determined by the above-mentioned price setting and order management unit 41. At this time, the prices and values ​​of the items included in the category of the calculation power ticket are added up to determine the price of the calculation power ticket.

[0084] For example, when a computing power user U who wants to use HW with high computing power immediately purchases a computing power ticket, he / she selects a high-performance item from among the items in the HW category. The price of this item varies depending on the HW performance, and generally the higher the performance, the more expensive it is. Also, even if the server has the same function, the price may differ depending on the HW vendor, etc.

[0085] Next, the reservation timing is the same day or the nearest date. The less flexibility there is in the date and time of the reservation, the more expensive it will be, and conversely, the further into the future the date, the lower the price may be. Finally, the calculation is scheduled to be performed in the time period closest to the reserved date and time. The more users want to use the computer during that time period, the more expensive the scheduled execution date will be. If the closest time period you want to use it is a busy time period, you will have to pay a higher price.

[0086] As a different example, when reserving computational power for a future date in units of several months, the purchase timing of the item included in the computational power ticket is several months in the future, so the computational power ticket can be purchased relatively cheaply. In addition, by avoiding the peak time period for the scheduled execution, it is possible to purchase a low-priced computational power ticket.

[0087] As a different example from the viewpoint of electricity, for example, a computational power user U who desires green electricity is assumed. When the execution of calculations using green electricity is assumed to be added value due to the business advantages of the computational power user U, it is assumed that the computational power user U will purchase computational power tickets with a higher price than general electricity. From this viewpoint, the computational power owner O sets the price of green electricity high and the price of general electricity low. Alternatively, when surplus electricity is used in green electricity, the price of electricity becomes lower than the price of general electricity. In that case, the price will be lowered by selecting green electricity.

[0088] In this way, the price is determined by adding up the prices of the items included in the calculation power ticket. At this time, the price of the item itself is determined by the provider of each item, which is equivalent to the cost price, and the calculation power owner O, who manages these items and categories, adds his / her own profit, etc. to determine the price of the calculation power ticket.

[0089] For example, the following are assumed to be factors that can be used to classify HW types and determine prices: Bare metal, VM (Virtual Machine), physical rental, etc. are assumed. For example, in the case of VM and bare metal, which can be used conveniently through communication, if it is easier for the computing power owner O to pre-configure bare metal than VM, bare metal can be provided at a lower cost. Also, for example, in the case of a computing power owner O that owns many types of machines, if costs can be reduced by configuring each machine as a single VM, VM can be provided at a lower cost. For example, the classification of HW configuration is assumed to include the CPU (vCPU) and the number of cores, clock speed, power consumption, heat generation, number of threads, architecture, etc.; the memory and its capacity, read / write speed, and cache configuration; the storage and its capacity, read / write speed, magnetic disk, flash memory, and raid settings; the GPU and its stream cores, clock speed, memory bandwidth, memory access speed, connection method such as PCIe, network connection method such as Ethernet and Infiniband, number of threads, architecture, power consumption, and heat generation; the line and its transmission / reception speed, compliant communication standard, cable length, contract provider, and VPN settings. The manufacturer, manufacturing date, manufacturing location, etc. of the above items are assumed, and the prices of each are calculated. However, not all of these categories are necessarily required. A calculation ticket may be made up of only some of these categories.

[0090] For example, in the case of SW classification, items that are assumed to be included include containers and the algorithms, libraries, programming languages, middleware, drivers, APIs, etc. that make up the containers. Items that are assumed to be included include OSs and the libraries, BIOS, kernels, programming languages, middleware, drivers, APIs, etc. that make up the containers. Items that are assumed to be included include drivers and the versions that make up the containers, compatible OSs, compatible HW, developers, etc. Items that are assumed to be included include analysis software and the versions that make up the containers, compatible OSs, compatible HW, developers, etc. Items that are assumed to be included include middleware and the versions that make up the containers, compatible OSs, compatible HW, developers, etc. The prices of each item are added up. However, not all of these categories are necessarily required. A calculation ticket may be made up of only some of these categories.

[0091] For example, in the case of data classification, items are assumed to include design data and its constituent dimensional data, material data, environmental data, operation data, etc.; analytical models and its constituent finite element models, particle models, trajectory models, probability models, distribution models, learning models, etc.; analytical conditions and its constituent input / output conditions, integration conditions, etc.; and the creator, creation date, academic citations, etc. of the above. The prices of each are added up. However, not all of these categories are necessarily required. A calculation ticket may be made up of only some of these categories.

[0092] For example, the following are assumed as elements that can be used to classify electricity and determine prices: general electricity, self-generated electricity, green electricity, surplus electricity, etc. For example, in the case of classification by installation location, items that can be considered include a large data center, the number of racks that make up the center, security level, lines, power consumption, support level, floor configuration, air conditioning, location, operating organization, etc. Items that can be considered include a container data center, the number of racks that make up the center, security level, lines, power consumption, support level, floor configuration, air conditioning, location, operating organization, etc. Items that can be considered include a server room, the number of racks that make up the center, security level, lines, power consumption, support level, floor configuration, air conditioning, location, operating organization, etc. The prices of each are added up. However, not all of these categories are necessarily required. A calculation ticket may be made up of only some of these categories.

[0093] For example, the following are assumed as factors that can be used to classify the type of computational user U and determine the price: Corporation, organization, academic, individual, and other conditions for computational user U (purpose of use of computational power, contract requirements, etc.). For example, factors that can be used to classify the timing of ticket purchase and determine the price include one month in advance, one week in advance, the day itself, and other conditions related to the use of computing power (conditions for determining whether it is used or unused, cancellation policy, etc.). Regarding the cancellation policy, possible circumstances that lead to cancellation include the circumstances of the computing power owner O, the circumstances of the computing power user U, and circumstances beyond one's control (natural disasters, etc.). For example, if the compensation due to the circumstances of the computing power owner O is generous, the price can be set high, and if the compensation is poor or nonexistent, the price can be set low. For example, the following are assumed as elements that can be used to classify the scheduled execution time slots and determine prices: weekday daytime (e.g., 9am-5pm), weekday nighttime (e.g., 5pm-10pm), weekday late night (e.g., 10pm-5am the next day), weekday early morning (e.g., 5am-9am), holiday daytime (e.g., 9am-5pm), holiday nighttime (e.g., 5pm-10pm), holiday late night (e.g., 10pm-5am the next day), holiday early morning (e.g., 5am-9am), calculation time: x hours (e.g., 6 hours out of 24 hours on any date), and other values ​​indicating time or duration are assumed.

[0094] For example, the following are assumed as factors that can be used to classify reliability and determine prices. Ratings and the computing power, communication speed, and security robustness of the servers that make up the ratings are assumed. Past performance and the percentage of servers that have been used in the past, and the computing and communication speeds achieved in the past are assumed. Security and the physical security of the servers that make up the ratings, security related to the software and virtual environment, and security related to communications are assumed. Uptime and the amount of time that the servers that make up the ratings continue to operate without power outages, and the percentage of servers that operate under appropriate temperature and humidity conditions are assumed. Owner information and the conditions defined by the individuals, corporations, research institutes, universities, and other computing power owners (O) that make up the ratings (purpose of use of computing power, contract requirements), etc. For example, the following are assumed as elements that can be used for dividing regions and determining prices: Tokyo (Japan), Osaka (Japan), Singapore, America, China, and other symbols or names indicating locations.

[0095] For example, the provider of an item is a power generation company for electricity, and the cost price is the price of electricity, etc. For HW, the provider is a HW vendor or agent such as a chip manufacturer or a system integrator, and the cost price is the price of the HW, etc. For other items, the cost price is the price paid by the computing power owner O when acquiring them.

[0096] If the computational power user U agrees to this price, the computational power ticket may be purchased at this price, but the computational power owner O may adjust the price set by the computational power user U to complete the transaction. This corresponds to the limit order and market transaction in the order management mentioned above. In addition to this, there may also be dynamic price fluctuations and static price setting methods. These are similar to current securities trading.

[0097] These computational power tickets may be purchased not only by users who use computational power, but also by third parties with the purpose of reselling the computational power tickets, which will be explained later as an example of an investor (see FIG. 16).

[0098] Next, the transition of components of calculation power in a calculation power ticket transaction will be described with reference to FIG. FIG. 10 is a diagram for explaining the transition of components of computational power in a computational power ticket transaction by dividing it into phases.

[0099] As shown in Figure 10, the components of computing power in computing power trading mainly transition through three phases: phase F1 of "provided computing power" in which computing power is provided by a computing power owner O, phase F2 of "purchased computing power" in which computing power is used by a computing power user U, and phase F3 of "resource management" in which resources are managed by the computing power owner O.

[0100] When the computing power owner O provides computing power to the server 1, the computing power is treated as provided computing power, and the server 1 enters the phase F1 of "provided computing power." In the "provided computing power" phase F1, when operation is performed using computing power coupons, computing power coupons are created and various computing power provided by various computing power owners O are managed as trading objects. For example, for power, computing power owner B provides power B, for servers, computing power owner C provides server C, and for VM source A, computing power owner A provides VM source A. The units of computing power provided and the definitions of computing power tickets are as described elsewhere, and computing power components other than those described in Fig. 10 are also managed as trading objects. In this example, the VM source and container are goods or information that combine software and data.

[0101] When a transaction of computing power is concluded for the provided computing power through an arbitrary computing power user U, a computing power usage contract is started at an appropriate time. For example, power B, server C, and VM source A can be purchased, and the phase transitions to "Purchased Computing Power" phase F2. In phase F2 of the "purchased computing power," the purchased computing power has the function of starting up VM source A in server C and starting the OS and analysis software that come with the VM source. In response to this, a computing power user U implements his / her own analysis data and executes the analysis to obtain the result data. During this process, power B is consumed.

[0102] After the usage contract ends, the system transitions to phase F3 of "resource management." In phase F3 of "resource management", for VM source A, configuration (repair, expansion) or recovery is performed by computing power owner A, who is the provider. For server C, configuration (repair, expansion) or recovery is performed by computing power owner C, who is the provider. For power B, configuration is performed by computing power owner B, who is the provider. At this point, if computing power owner O recovers or does not confirm the configuration, it will no longer be treated as provided computing power and will not be subject to trading. By confirming the configuration and providing it as computing power, it will become subject to trading as provided computing power. The above three-phase cycle involves the provision, utilization, and resource management of computing power.

[0103] A server according to a second embodiment of the information processing device of the present invention will be described. First, the functional configuration of the server of the second embodiment will be described with reference to FIG. FIG. 11 is a functional block diagram showing a second embodiment of the functional configuration of a server having the hardware configuration of FIG. In describing the second embodiment, the same components as those in the first embodiment are given the same reference numerals and description thereof will be omitted. As shown in FIG. 11, in the CPU 11 of the server 1, a selling price management unit 30, a computing power management unit 50, and a token management unit 70 function when executing processing related to buying and selling transactions of computing power. That is, FIG. 11 shows the server 1 of the second embodiment including an additional functional configuration, in which a token management unit 70 is further added to the selling price management unit 30 and the computing power management unit 50 shown in FIG.

[0104] The selling price management unit 30 comprises a price setting and order management unit 41 , a cash conversion unit 42 , and a fund management unit 43 .

[0105] The cashing unit 42 exchanges the value of the computational power ticket that has been won by the computational power user U into cash or the like paid by the fund management unit 43.

[0106] The token management unit 70 includes a point management unit 81 and a ticket management / matching unit 82 .

[0107] The point management unit 81 issues a separately determined amount of points according to the amount of money input by the computationally capable user U. The issued points are sent to the computationally capable user U. In this way, by introducing points as a value used in transactions and a points management unit 81 that manages the points, the net value of computing power can be evaluated without the influence of exchange rates, as compared to the case of using cash. For example, the value of a computing resource such as a single server as computing power can be quantified independently of the price fluctuation of JPY / USD. In addition, the influence of the transaction currency of the computing power owner O and the computing power user U is reduced, and transactions can be carried out smoothly even if the computing power owner O and the computing power user U use different currencies.

[0108] The ticket management and matching unit 82 manages the computing power tickets issued in response to the computing power provided by the computing power owner O. Specifically, the ticket management and matching unit 82 searches whether the desired computational power ticket exists at a given time, and if the search results in the existence of multiple computational power tickets that satisfy the constraints (purchase conditions) of the computational power user U, it performs a matching process between the constraints of the computational power user U and the computational power ticket (contents of computational power). Matching is performed by at least one of the following methods: a method of uniquely determining one ticket for multiple tickets based on a specific computational ability component as an index; a method of calculating one ticket that is most advantageous to the computational ability user U, such as having a low selling price; a method of disclosing all applicable tickets to the computational ability user U who then selects one ticket; and a method of performing a recommendation evaluation on all applicable tickets to encourage the computational ability user U to select one ticket. The matching algorithm may be selected by the ticket management and matching unit 82, which may receive multiple algorithms from the user U of computing power. By providing multiple different matching algorithms, the principle of competition comes into play, and the computing power market is optimized. In the above, the various algorithms can treat the components of the computational power written on the computational power ticket as data, and can also input constraints of the computational power user U as necessary. Although an example of a case where there are multiple tickets with the same content has been described above, any method can be used to select one ticket, and for example, one ticket from multiple tickets may be matched in the order provided by the computational power owner O. When a match is made, the computational power ticket is traded. After the sale and purchase of the computational power ticket is completed, the ticket management and matching unit 82 can acquire (obtain) the computational power ticket by paying for the computational power ticket using the points already acquired and inserted into the point management unit 81 by the computational power user U. The ticket management and matching unit 82 notifies the computing power receipt management unit 62 of information (e.g., payment has been made) on the computing power ticket acquired (obtained) by the computing power user U. This allows the computing power user U to receive computing power from the computing power receipt management unit 62 or the computing power owner O.

[0109] According to the configuration of the server 1 of the second embodiment, by adding a token management unit 70 in addition to the functional configuration of the server 1 of the first embodiment shown in Figure 4, it is possible to improve the convenience, robustness, and soundness of computational power trading.

[0110] Next, the operation of the server 1 of the second embodiment will be described with reference to FIG. The operation of the server 1 will be described in detail with reference to FIG. 12, which shows an example of an operational flow of each of the above-mentioned components when the server 1 in FIG. 11 is operated. FIG. 12 is a diagram for explaining the flow of processing of computational power trading in a server having the functional configuration (second embodiment) of FIG.

[0111] The processing of computational power trading in the server of the second embodiment includes the following steps S31 to S36. 12, when information on computing power is provided from the computing power owner O, the computing power provision management unit 61 manages the provided information on computing power. Here, providing the information on computing power means making the components of the computing power available to the computing power user U, and may be making the components of the computing power available to the computing power user U at a future date and time or for a future period of time by contract or the like. The computing power provision management unit 61 receives the content of the provision of computing power (information on computing power) by the computing power owner O. The received content is stored in the DB 90 and shared with the ticket management and matching unit 82. Here, sharing means that the ticket management and matching unit 82 is notified by the computing power provision management unit 61 that computing power information has been provided by the computing power owner O, and the contents are made available to be read from the DB 90 at any time. At this time, the computing power owner O may accept sales conditions such as a desired selling price of the computing power. The ticket management and matching unit 82 issues a ticket indicating the shared computational power provision content. The ticket is preferably electronic. Ticket issuance information is shared with the price setting and order management unit 41. The ticket may include a desired selling price by the computational power owner O.

[0112] In step S32 (fund input step), the computationally skilled user U inputs funds into the fund management unit 43. The fund management unit 43 accepts a contract for sending or receiving cash or money and the amount. The accepted amount is shared with the point management unit 81. The point management unit 81 issues a separately determined amount of points according to the amount of money. The issued points are sent to the computationally skilled user U.

[0113] In step S33 (market buying and selling step of computational power tickets), after the ticket issuance information is shared from the ticket management and matching unit 82 to the price setting and order management unit 41, the price setting and order management unit 41 determines the selling price of the ticket in points (hereinafter referred to as the point selling price) in a predetermined manner. The selling price of points may be, for example, a fixed absolute price determined in advance based on the quantitative and qualitative evaluation of the calculation power written on the ticket, or may be determined relatively by comparing multiple calculation powers. In addition, the price may be determined based on market principles by limit orders and market orders in accordance with the method of stock market trading. In this case, the calculation power user U may apply various orders by setting restrictions such as minimum conditions and desired conditions for the components of calculation power. For example, the price setting and order management unit 41 discloses the contents of the ticket to a computational user U, accepts the computational user U's quantitative and qualitative evaluation of the ticket in the form of a point selling price, and at the same time accepts a desired selling price from a computational owner O, and executes various orders based on a comparison of the two selling prices. For example, if a computational user U places a limit order for a ticket with certain contents at 100 points, and the computational owner O for that ticket has set a desired selling price of 90 points or more, a trade will be made at 100 points. Alternatively, if a computational user U places a market order for two tickets with the same contents, and the computational owner O for one ticket has set a desired selling price of 90 points or more and the computational owner O for the other ticket has set a desired selling price of 80 points or more, a trade will be made at 80 points. A computationally capable user U can set constraints (conditions) such as minimum conditions and desired conditions for a desired ticket. Whether or not a desired ticket exists at any time is searched for by the ticket management and matching unit 82. If the search results in the existence of multiple tickets that satisfy the constraints of the computationally capable user U, matching is performed between the constraints of the computationally capable user U and the tickets. Matching can be performed by a method of uniquely determining one ticket for multiple tickets based on a specific computational power component as an index, a method of calculating one ticket that is most advantageous to the computational power user U, such as having a low selling price, a method of disclosing all applicable items to the computational power user U and having the computational power user U select one ticket, a method of performing a recommendation evaluation of all applicable items to encourage the computational power user U to select one ticket, etc. The matching algorithm may be separately accepted by the ticket management and matching unit 82, and the computational ability user U may select from multiple algorithms. By providing multiple different matching algorithms, the principle of competition comes into play and the computational ability market is optimized. In the above, various algorithms can treat the components of the computational ability written on the ticket as data, and can also enter constraints on the computational ability user U as necessary. When there are multiple tickets with the same content, it is sufficient to select one ticket in any manner. For example, one ticket can be matched from the multiple tickets in the order provided by the computing power owner O. If a match is made, the calculation ticket is traded.

[0114] In step S34 (bidding step for the calculation power ticket), after the sale of the calculation power ticket is completed, the calculation power user U inputs the points he or she has already acquired into the point management unit 81, and purchases and acquires (obtains) the ticket via the ticket management / matching unit 82. In addition, the computing power owner O obtains points from the point management unit 81 based on the matching results established by the ticket management and matching unit 82.

[0115] In step S35 (step of receiving computing power), the computing power user U can receive computing power from the computing power receiving management unit 62 by inserting the computing power ticket into the ticket management and matching unit 82. This allows the computational power user U to perform calculations. If necessary, computational power may be provided directly from the computational power owner O. In that case, the ticket will state that computational power will be provided directly from the computational power owner O.

[0116] In step S36 (point conversion step), when the computing power owner O inputs points into the server 1, the point management unit 81 in the server 1 associates the input points with the computing power owner O's account, stores the input points in the DB 90, and shares the points with the conversion unit 42. The cash conversion unit 42 converts the shared points into cash or the like via the fund management unit 43, and pays the computational power owner O the value of the computational power purchased by the computational power user U.

[0117] As described above, according to the server 1 of the second embodiment, the server 1 can realize a series of service flows, such as provision of computing power from a computing power owner O, receiving computing power from a computing power user U, accepting payment from the computing power user U, and paying the payment to the computing power owner O. By introducing points and the point management unit 81, the net value of computing power can be evaluated without being affected by exchange rates, as compared to using cash. For example, the value of the computing resources of one server can be quantified independently of the price fluctuation of JPY / USD. By introducing the computational power ticket, multiple indicators that characterize computational power (electricity, hardware, software, data, etc.) can be listed in one unit. For example, if the specifications and location of a server are included in the indicators in the computational power ticket, and two arbitrary servers (computational power tickets) have different specifications and locations but have the same overall computing power, the computational power ticket can show an indicator that is sufficient for judgment to a computational power user U who prioritizes specifications and a computational power user U who prioritizes location.

[0118] The ticket management and matching unit 82 allows the computational power user U to easily issue a desired ticket, i.e., computational power. The computational power of the computational power owner O, which was previously unknown, can be used. In addition, computational power can be provided from the computational power owner O to the computational power user U at a selling price based on the balance of supply and demand of computational power.

[0119] The introduction of computational power tickets allows transactions to be conducted for each component of computational power. This improves the convenience of computing resource sharing for both computational power owners O and computational power users U. In addition, as mentioned above, the price determination method based on market principles improves the soundness of computational power trading.

[0120] Below, we will explain the specific flow of computing power trading using several examples. First, a simple example of the flow of computational power trading will be described with reference to FIG. FIG. 13 is a diagram showing an example of a simple flow of computational power trading in the information processing system shown in FIG. 13, m computing power owners O (m is an integer value of 1 or more) who possess computing power input resource information that can be provided from their own computing power to the server 1. When inputting information, multiple elements such as the installation location, security level, whether or not green power is used, hardware configuration, software, and data are included, and the number of points according to each specification is defined. n users U (n is an integer value of 1 or more) of computing power who use computing power can select, reserve, and use resources that meet their requirements from the resource information published on the server 1. It is possible to register and reserve resources based on information obtained not only from the computing power users U and computing power owners O themselves, but also from the exchange operating company that operates and manages this device.

[0121] Next, an example of computing power trading in insurance services will be described with reference to FIG. FIG. 14 is a diagram showing an example of computing power trading in insurance services in the information processing system shown in FIG. In the example of Fig. 14, an insurance computing power provider H is added. When the demand of a computing power user U exceeds the amount of resources provided by the computing power owner O or when there are no resources that meet the conditions, the insurance computing power provider H obtains information on the shortage of computing power from the server 1 and compensates for the computing power. The method and content of compensation may be anything that can be agreed upon with the computational power user U, and may be, for example, providing separate computational power (e.g., a server) with specifications equivalent to those desired by the computational power user U, providing separate computational power (e.g., electricity) of the same value as the computational power (e.g., a server) desired by the computational power user U, returning points (tokens) or money equivalent to the specifications desired by the computational power user U, etc.

[0122] Next, with reference to FIG. 15, an example in which a rating function for rating the computational resources provided by computational resource owners is added will be described. FIG. 15 is a diagram showing an example in which a rating function for rating the computational resources provided by computational resource owners is added. 15, the rating server 5 has a rating function that evaluates the computational resources provided by the computational resource owner O. The rating function is realized by executing a thinking program, such as an artificial intelligence (AI), installed in the rating server 5, for example. The rating function objectively determines the market value based on the resource information provided by the computing power owner O, and determines the price to be offered to the computing power user U. In addition, the content of the rating can be reflected in the computing power ticket, and can function as reference information when the computing power user U selects a computing power ticket. This allows the computational user U to use computational resources with a number of points according to the market price.

[0123] Next, with reference to FIG. 16, an example in which an investor element is further added to the example of FIG. 15 will be described. FIG. 16 is a diagram showing an example in which an element of an investor is further added to the example of FIG. In the example of FIG. 16, an element of investor T is added to the elements of FIG. In this case, each computational power ticket is assigned a number of points evaluated by the rating function, but since the number of points fluctuates depending on demand and market trends, it is possible to purchase computational power tickets in advance when the number of points required is low, and sell computational power tickets when demand, i.e., the number of points, is high. The decision to purchase computational power tickets based on point fluctuations applies not only to investors T but also to computational power users U.

[0124] Although one embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiment, and modifications, improvements, etc. within the scope that can achieve the object of the present invention are included in the present invention.

[0125] Furthermore, for example, the series of processes described above can be executed by hardware or software. In other words, the functional configurations in FIG. 4 and FIG. 11 are merely examples and are not particularly limited. That is, it is sufficient that the information processing system or information processing device is provided with a function capable of executing the above-mentioned series of processes as a whole, and the type of functional block used to realize this function is not limited to the examples of Figures 4 and 11. Furthermore, the locations of the functional blocks and databases are not particularly limited to those of Figures 4 and 11, and may be arbitrary. For example, at least a part of the functional blocks and databases required for executing various processes may be transferred to a user terminal or the like. Conversely, the functional blocks and databases of the user terminal may be transferred to a server or the like. Furthermore, one functional block may be configured as a single piece of hardware, a single piece of software, or a combination of both.

[0126] When the series of processes is executed by software, the program constituting the software is installed into a computer or the like from a network or a recording medium. The computer may be a computer implemented with dedicated hardware. Furthermore, the computer may be a computer capable of executing various functions by installing various programs thereon, such as a server, a general-purpose smartphone, or a personal computer.

[0127] A recording medium containing such a program may be composed not only of a removable medium (not shown) that is distributed separately from the device main body in order to provide the program to a user, etc., but also of a recording medium that is provided to a user, etc. in a state where it is pre-installed in the device main body.

[0128] In this specification, the steps of describing a program to be recorded on a recording medium include not only processes that are performed chronologically according to the order, but also processes that are not necessarily performed chronologically but are executed in parallel or individually. In addition, in this specification, the term "system" refers to an overall device that is composed of a plurality of devices, a plurality of means, etc.

[0129] In other words, it is sufficient for an information processing device to which the present invention is applied to have the following configuration, and various embodiments can be adopted. That is, an information processing device to which the present invention is applied (for example, the server 1 in FIG. 4) (1) A computing power value evaluation means (e.g., the price setting and order management unit 41 of the selling price management unit 30 in FIG. 4) for evaluating a predetermined value (e.g., XX points, etc.) for a right to use a predetermined unit of computing power (e.g., a computing power ticket) defined by a plurality of indices (e.g., power, hardware, software, data) provided by a computing power owner (e.g., the computing power owner O in FIG. 4), Equipped with. This makes it possible to improve the convenience of providing, using, and managing computing resources.

[0130] (2) a ticket issuing and transferring means (price setting and order management unit 41 of selling price management unit 30 in FIG. 4) which issues a ticket for exercising the right (e.g., computing power tickets 91 and 92 in FIG. 8 and FIG. 9) to the computing power owner (e.g., computing power owner O in FIG. 4) and executes control to transfer the ticket (e.g., computing power tickets 91 and 92 in FIG. 8 and FIG. 9) from the computing power owner (e.g., computing power owner O in FIG. 4) to a computing power user (e.g., computing power user U in FIG. 4) at the predetermined value (e.g., points); a computing power management means (e.g., the computing power management unit 50 in FIG. 4) for receiving the predetermined unit of computing power indicated on the ticket (e.g., the computing power ticket 91, 92 in FIG. 8, FIG. 9) from the computing power owner (e.g., the computing power owner O in FIG. 4) to whom the ticket (e.g., the computing power ticket 91, 92 in FIG. 8, FIG. 9) has been issued, and for executing control to cause the computing power user (e.g., the computing power user U in FIG. 4) to whom the ticket (e.g., the computing power ticket 91, 92 in FIG. 8, FIG. 9) has been transferred to receive the predetermined unit of computing power indicated on the ticket (e.g., the computing power ticket 91, 92 in FIG. 8, FIG. 9); Further comprising: It is possible.

[0131] (3) When a plurality of tickets (e.g., the computing power management unit 50 in FIG. 4) having at least some of the same indicators are used, the computing power management means (e.g., the computing power management unit 50 in FIG. 4) executes control to allow the computing power user (e.g., the computing power user U in FIG. 4) to receive the predetermined unit of computing power by giving priority to the ticket (e.g., the computing power ticket 91, 92 in FIG. 8, 9) having a higher predetermined priority (e.g., in order of earliest resource registration of the computing power owner O) among the plurality of tickets (e.g., the computing power ticket 91, 92 in FIG. 8, 9). It is possible.

[0132] (4) A matching means (e.g., the ticket management and matching unit 82 in FIG. 11) is further provided for executing control to put up a predetermined ticket (e.g., the calculation power ticket 91, 92 in FIG. 8 and FIG. 9) and to make the ticket (e.g., the calculation power ticket 91, 92 in FIG. 8 and FIG. 9) a successful bid for a calculation power user among a plurality of the calculation power users (e.g., the calculation power user U in FIG. 4) who satisfies a predetermined successful bid condition, The calculation power value evaluation means (e.g., the price setting and order management unit 41 of the selling price management unit 30 in FIG. 4) evaluates the value of the successful bid ticket (e.g., the calculation power ticket 91, 92 in FIG. 8 and FIG. 9) based on the specified bidding conditions. As a result, the value at the time of evaluation will be reflected in the evaluation of new computing power provided by computing power owners in the future.

[0133] (5) The computational power value evaluation means (e.g., the price setting and order management unit 41 of the selling price management unit 30 in FIG. 4) selects the ticket (e.g., the computational power ticket 91, 92 in FIG. 8 and FIG. 9) that is suitable for the conditions specified by the computational power user (e.g., the computational power user U in FIG. 4) and recommends the ticket (e.g., the computational power ticket 91, 92 in FIG. 8 and FIG. 9) to the computational power user (e.g., the computational power user U in FIG. 4); This allows a computational power user to easily find a ticket (e.g., a computational power ticket) containing the desired content, without having to check the content of each ticket (e.g., a computational power ticket) that has slightly different computational power content, among the many tickets (e.g., computational power tickets) being offered for sale.

[0134] (6) The computing power management means (e.g., the computing power management unit 50 in FIG. 4) provides computing power to the computing power user (e.g., the computing power user U in FIG. 4) that meets the conditions specified by the computing power owner (e.g., the computing power owner O); It is possible.

[0135] (7) In the ticket issuing and transferring means (the price setting and order management unit 41 of the selling price management unit 30 in FIG. 4) and the computing power management means (for example, the computing power management unit 50 in FIG. 4), a token such as a point is used as a transaction unit. It is possible.

[0136] (8) An information processing method to which the present invention is applied comprises: An information processing method executed by an information processing device, A step of evaluating a predetermined value (e.g., XX points, etc.) for a right to use a predetermined unit of computing power (e.g., computing power coupons 91 and 92 in Figs. 8 and 9) defined by a plurality of indicators (e.g., power, hardware, software, data) provided by a computing power owner (e.g., computing power owner O in Fig. 4); Including, It is possible.

[0137] (9) The program to which the present invention is applied is On the computer, A step of evaluating a predetermined value (e.g., XX points, etc.) for a right to use a predetermined unit of computing power (e.g., computing power coupons 91 and 92 in Figs. 8 and 9) defined by a plurality of indicators (e.g., power, hardware, software, data) provided by a computing power owner (e.g., computing power owner O in Fig. 4); Executing a control process including It is possible. [Explanation of symbols]

[0138] 1: Server, 2: User terminal, 11: CPU, 12: ROM, 13: RAM, 14: Bus, 15: Input / output interface, 16: Input section, 17: Output section, 18: Memory section, 19: Communication section, 20: Drive, 21: Removable media, 30: Selling price management section, 41: Price setting and order management section, 42: Cash conversion section, 43: Funds management section, 50: Computing power management section, 61: Computing power provision management section, 62: Computing power receipt management section, 70: Token management section, 81: Points management section, 82: Ticket management and matching section, S: Service manager, NW: Network, 90: Database (DB), 91, 92: Computing power ticket

Claims

1. a computing power value evaluation means for evaluating a predetermined value for a right to use a predetermined unit of computing power defined by a plurality of indexes provided by a computing power owner; a ticket issuing and transferring means for issuing a ticket for exercising the right to the computing power owner and executing control for transferring the ticket from the computing power owner to a computing power user at the predetermined value; a computing power management means for receiving the predetermined unit of computing power indicated on the ticket from the computing power owner to whom the ticket has been issued, and for executing control to allow the computing power user to whom the ticket has been transferred to receive the predetermined unit of computing power indicated on the ticket; An information processing device comprising:

2. When a plurality of tickets having at least some of the plurality of indices that are the same are used, the computing power management means executes control to cause the computing power user to receive the predetermined unit of computing power by giving priority to a ticket having a higher predetermined priority among the plurality of tickets. The information processing device according to claim 1 .

3. A matching means is further provided for executing control to put up a predetermined ticket and to make a bid for the predetermined ticket to a user among the plurality of users with computing ability who satisfies a predetermined bidding condition, The calculation power value evaluation means evaluates the value of the successful bid certificate based on the predetermined bidding conditions. The information processing device according to claim 1 .

4. The computing power value evaluation means selects the ticket that meets the conditions specified by the computing power user, and recommends the ticket to the computing power user. The information processing device according to claim 1 .

5. The computing power management means provides computing power to the computing power user that meets the conditions specified by the computing power owner. The information processing device according to claim 1 .

6. The certificate issuing and transferring means and the computing power managing means use a token as a transaction unit. The information processing device according to claim 1 .

7. An information processing method executed by an information processing device, evaluating a predetermined value for a right to use a predetermined unit of computing power defined by a plurality of indices provided by a computing power owner; a ticket issuing and transferring step of issuing a ticket for exercising the right to the computing power owner and executing control to transfer the ticket from the computing power owner to the computing power user at the predetermined value; a computing power management step of receiving the predetermined unit of computing power indicated on the ticket from the computing power owner to whom the ticket has been issued, and executing control to allow the computing power user to whom the ticket has been transferred to receive the predetermined unit of computing power indicated on the ticket; An information processing method comprising:

8. On the computer, evaluating a predetermined value for a right to use a predetermined unit of computing power defined by a plurality of indices provided by a computing power owner; a ticket issuing and transferring step of issuing a ticket for exercising the right to the computing power owner and executing control to transfer the ticket from the computing power owner to the computing power user at the predetermined value; a computing power management step of receiving the predetermined unit of computing power indicated on the ticket from the computing power owner to whom the ticket has been issued, and executing control to allow the computing power user to whom the ticket has been transferred to receive the predetermined unit of computing power indicated on the ticket; A program that executes control processing including:

Images