JP2007249760A - Distributed mutual backup system, information processor, distributed mutual backup method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a distributed mutual backup system making cooperative operation performing effective use a hardware resource of each computer be performed by heightening probability that the computer storing backup data is usable when needing a backup. <P>SOLUTION: A backup data production means 102 produces the backup data to be distributed to the other computers from data produced by a user. A terminal use prediction means 103 communicates with terminal use prediction means 103 of the other computers 100-2 to 100-n through a data communication means 104, collects data on a state of each the computer, determines a backup destination, and instructs the backup destination to the backup data production means 102. A backup data storage means 105 stores the backup data sent from data communication parts 104 of the other computers 100-2 to 100-n. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a distributed mutual backup system, an information processing apparatus, a distributed mutual backup method, and a program in which a plurality of computers each store backup data in another computer.

  In companies and the like, a large number of personal computers are used, and data storage devices (hard disks and the like) built in each computer store (store) data such as documents created by users.

  In preparation for a case where a data storage device built in each computer fails and data stored in each computer is lost, backup data is created and stored. There are the following methods for storing the backup data.

  (1) A method of storing backup data in a removable recording medium such as a CD-R (Compact Disk Recordable) or a DVD-R (Digital Versatile Disk Recordable).

  (2) A method of storing backup data in an unused area of the same data storage device or in another data storage device built in the same computer.

  (3) A method of transmitting backup data to a data backup server via a communication network and storing the backup data in the server.

  In general, since the storage capacity of a removable recording medium is smaller than the storage capacity of a data storage device built in the computer, a plurality of storage media are required to store backup data in the storage medium. Therefore, the method shown in (1) requires a complicated operation of storing backup data in a plurality of storage media and managing the plurality of storage media storing the backup data. Therefore, there is a problem that it becomes difficult to frequently store backup data.

  Further, the method shown in (2) has a problem that when backup data is stored in an unused area in the data storage device, the backup data is lost if a failure occurs in the data storage device itself. Further, when backup data is stored in another data storage device built in the same computer, there is a problem that backup data is lost when a common part such as a power source of the computer fails.

  Further, the method shown in (3) has a problem that the server needs to have a large-capacity data storage device.

  Therefore, Patent Document 1 describes a system in which a plurality of computers divide backup data and store them in a data storage device provided in each computer.

  The system described in Patent Document 1 will be described. FIG. 14 is a block diagram showing a system described in Patent Document 1. As shown in FIG. The system shown in FIG. 14 includes a plurality of information processing apparatuses 20-1 to 20-10 connected to each other via a communication network and a router.

  FIG. 15 is a block diagram illustrating a configuration of an information processing apparatus that configures the system described in Patent Document 1 illustrated in FIG. 14. The information processing apparatus 20-1 illustrated in FIG. 15 includes a storage unit 200, a dividing unit 210, and a backup unit 270. Although FIG. 15 shows the information processing apparatus 20-1 as an example, the other information processing apparatus shown in FIG. 14 has the same configuration.

  The storage unit 200 stores data used by the information processing apparatus. The dividing unit 210 divides the data stored in the storage unit 200 into a plurality of group data and outputs it to the backup unit 270. The backup unit 270 transmits each of the plurality of group data divided and output by the dividing unit 210 to each of the other plurality of information processing apparatuses and stores them.

  Therefore, the system described in Patent Document 1 allows the free capacity of the data storage devices of a plurality of computers to be interchanged, and the backup data of the own computer is divided and stored in a plurality of other computers via a communication network. Therefore, the above problems (1) to (3) are solved.

  Patent Document 2 describes a system that stores backup data of one server in another server ranked by a predetermined method.

JP 2005-115524 A (paragraphs 0016 to 0079, FIG. 1) JP 2002-215474 A (paragraphs 0024 to 0047, FIG. 1)

  The systems described in Patent Document 1 and Patent Document 2 have a problem that when backup data is necessary, the data may not be used. The reason is that whether or not a personal computer such as a company is powered on is left to the discretion of the computer user, and the working hours of the computer user who stores the backup data This is because the backup data cannot be used because the computer is not in the power-on state when the backup data is required.

  Moreover, the system described in Patent Document 1 has a problem that it does not have a mechanism for efficiently using hardware resources included in each information processing apparatus.

  Therefore, when the backup destination information processing apparatus for storing backup data is determined randomly or in the order of IP addresses, the problem (1) cannot be solved.

  In addition, when the user can freely select the backup destination, the user wants to store the backup data in the information processing apparatus that has been operating for a longer period of time and operates only for a short period of time. It is considered that it is not desired to store backup data in an information processing apparatus that does not exist.

  Furthermore, when an information processing device that has been operating for only a short time stores backup data in another information processing device, each information item that has no penalty for the information processing device that has been operating for only a short time A user of a processing apparatus desires to store backup data in an information processing apparatus that operates for a long time regardless of whether the operating time of the apparatus is long or short. Therefore, the allocation of the backup data storage according to the startup time of the information processing apparatus and the motivation for making it easy for other users to perform the backup are not performed. Therefore, as a result, an imbalance arises in the role and responsibility of each information processing apparatus.

  Therefore, the present invention increases the probability that a computer storing backup data can be used when using backup data, and increases the certainty of backup. It is an object to provide a backup method and program. Another object of the present invention is to provide a distributed mutual backup system, an information processing apparatus, a distributed mutual backup method, and a program that can effectively use hardware resources of each computer.

  The distributed mutual backup system according to the present invention is a distributed mutual backup system in which a plurality of information processing devices transmit and receive backup data to each other via a communication network, and each information processing device creates backup data of stored data Backup data creation means that predicts the operating time of its own device and the operating time of another information processing device, and determines the backup data based on the predicted operating time of its own device and the predicted operating time of the other information processing device. Terminal usage prediction means for selecting an information processing apparatus as a transmission destination of backup data for storage.

  The terminal usage prediction means obtains a correlation that is a ratio of overlap between the predicted operation time of the own device and the predicted operation time of another information processing device, and determines an information processing device having a high correlation of the predicted operation time as a transmission destination of backup data. You may choose as

  Based on the correlation between the predicted operation time of the own device and another information processing device, the terminal usage prediction means sends a predetermined number of information processing devices to the backup data transmission destination in descending order of the probability of operation at the same time. You may choose as

  Each information processing device includes a backup data storage unit that stores backup data of another information processing device, a price setting unit that sets a price for storing backup data of another information processing device in the backup data storage unit, Another information processing apparatus may include budget setting means for setting a budget for transmitting and storing backup data.

  An information processing apparatus according to the present invention is an information processing apparatus that is mutually connected via a communication network and transmits / receives backup data, and includes backup data creation means for creating backup data of stored data, The backup data transmission destination to store the backup data based on the predicted operation time of the local device and the predicted operation time of the other information processing device. Terminal usage predicting means for selecting an information processing apparatus.

  The distributed mutual backup method according to the present invention is a distributed mutual backup method in which a plurality of information processing apparatuses connected via a communication network transmit / receive backup data to / from each other. Predict the operating time of the device and the operating time of the other information processing device, and store the backup data of the stored data based on the predicted operating time of the own device and the predicted operating time of the other information processing device A terminal usage prediction step for selecting an information processing device as a transmission destination of backup data, and a backup data creation step for creating backup data for stored data in the backup data creation means of the information processing device. It is characterized by.

  The terminal usage prediction unit of the information processing device obtains a correlation that is a ratio of overlap between the predicted operation time of the own device and the predicted operation time of another information processing device in the terminal usage prediction step, and the correlation of the predicted operation time is high. The information processing apparatus may select backup data as a transmission destination.

  The terminal usage predicting means of the information processing device is a terminal usage prediction step, based on the correlation of the predicted operating time between the own device and another information processing device, in a descending order of the probability of operating at the same time. May be selected as the backup data transmission destination.

  The data backup unit of the information processing device stores the backup data of the other information processing device in the backup data storage unit, and the price setting unit of the information processing device stores the backup data of the other information processing device. A price setting step for setting a price when storing the data in the backup data storage means, and a budget for setting a budget when the budget setting means of the information processing apparatus transmits and stores the backup data in another information processing apparatus A setting step.

  A distributed mutual backup program according to the present invention is a distributed mutual backup program that is mounted on an information processing apparatus that is mutually connected via a communication network and transmits / receives backup data. Backup data transmission destination to predict the operation time of the information processing device and store the backup data of the stored data based on the predicted operation time of the own device and the predicted operation time of the other information processing device A terminal usage prediction process for selecting an information processing apparatus to be used, and a backup data creation process for creating backup data of stored data.

  In the terminal usage prediction process, the computer obtains a correlation that is a ratio of the overlap between the predicted operating time of the own device and the predicted operating time of another information processing device, and backs up the information processing device having a high correlation of the predicted operating time to the backup data. It may be selected as a transmission destination.

  In a terminal usage prediction process, a predetermined number of information processing devices are backed up in descending order of the probability of operating at the same time based on the correlation of the predicted operation time between the own device and another information processing device. May be selected by the information processing apparatus that transmits the message.

  Backup data storage process for storing backup data of other information processing apparatus in backup data storage means in computer, and price setting process for setting price when storing backup data of other information processing apparatus in backup data storage means Also, another information processing apparatus may execute budget setting processing for setting a budget for transmitting and storing backup data.

  ADVANTAGE OF THE INVENTION According to this invention, when backup data is required, the probability that the information processing apparatus which stores backup data can be utilized can be raised. In addition, hardware resources included in each information processing apparatus can be effectively used. The reason for this is to select an information processing apparatus to which backup data is to be transmitted based on prediction of operating time between the own apparatus and another information processing apparatus.

  The terminal usage prediction means is configured to select an information processing device that transmits backup data based on a correlation that is a ratio of an overlap between the predicted operation time of the own device and the predicted operation time of another information processing device. In this case, the entire system can be operated in a coordinated manner to increase the certainty of backup and to effectively use the hardware resources included in each information processing apparatus.

  The information processing apparatus includes price setting means for setting a price for storing backup data of another information processing apparatus, and budget setting means for setting a budget for storing backup data in the other information processing apparatus In such a case, the acceptance of backup data can be controlled by setting a budget and price.

Embodiment 1 FIG.
A first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a first embodiment of a distributed mutual backup system according to the present invention.

  The distributed mutual backup system shown in FIG. 1 includes computers (information processing apparatuses) 100-1 to 100-n that are connected to each other by a communication network and operate according to program control.

  A computer 100-1 shown in FIG. 1 includes user data storage means 101, backup data creation means 102, terminal usage prediction means 103, data communication means 104, and backup data storage means 105. Note that each of the computers 100-2 to 100-n illustrated in FIG. 1 has the same configuration as the computer 100-1.

  The user data storage unit 101 stores (stores) all data created by the user using the computer 100-1.

  The backup data creation unit 102 creates data (hereinafter referred to as backup data) for copying data created by the user and storing it in another computer. When the data created by the user is duplicated, the backup data creation means 102 may simply duplicate the data created by the user, or a RAID (Redundant Arrays of Inexpensive) used in a normal file system. Similarly to the Disks technology, error correction code data may be created, and backup data distributed to a plurality of other computers may be created in a form including the error correction code data.

  The backup data creation unit 102 restores the backup data received by the data communication unit 104 from another computer.

  The terminal usage predicting means 103 predicts the operating time of the computer 100-1 based on past operating results and the like.

  The terminal usage prediction means 103 communicates with the terminal usage prediction means 103 of the other computers 100-2 to 100-n via the data communication means 104, and collects information on the state of each computer. The terminal usage prediction unit 103 predicts the operating time of another computer based on the collected information. Then, the terminal usage prediction unit 103 determines the transmission destination of the backup data based on the predicted result, and notifies the backup data creation unit 102 of the determined transmission destination.

  The data communication means (data backup means) 104 manages communication between terminal prediction means 103 of computers and transmission / reception of backup data between computers. The backup data storage unit 105 stores (stores) backup data received from the data communication unit 104 of the other computers 100-2 to 100-n.

  Each computer predicts the operating time of its own device and the operating time of another computer in the computer, and stores it based on the predicted operating time of its own device and the predicted operating time of the other computer. Distributed mutual backup for executing terminal usage prediction processing for selecting a backup data transmission destination computer for storing backup data and backup data creation processing for creating backup data for stored data A program may be installed.

  Next, the operation of the first exemplary embodiment of the present invention will be described with reference to the drawings. FIG. 2 is a sequence diagram for explaining the operation when the computer 100-1 transmits backup data to the computer 100-2.

  The terminal usage predicting unit 103 includes a data or a time interval at which the data to be backed up among the data stored in the user data storage unit 101 and the backup data of the data are transmitted to another computer by the user. Is designated (step S101). The terminal usage predicting unit 103 communicates with the terminal usage predicting unit 103 of another computer via the data communication unit 104 to transmit information on the operating time to the other computer (step S102), and the operating time of the other computer Is collected (step S103).

  The terminal usage predicting means 103 is such that the computer 101-1 operates at the time until the backup data designated by the user is transmitted to another computer and the time until the next backup data is transmitted to the other computer. And the operation time of the other computer based on the information of the operation time of the other computer collected through the data communication unit 104 is performed (step S104).

  Based on the predicted result, the terminal usage predicting means 103 selects a plurality of computers having operating times that are highly common in the operating time of the computer 101-1 as backup data transmission destination candidates (step S 105). .

  The terminal usage prediction unit 103 confirms whether the computer selected via the data communication unit 104 can accept the backup data (step S106), and determines the transmission destination of the backup data. Specifically, the terminal usage prediction unit 103 inquires of the computer selected as the backup data transmission destination candidate whether the backup data may be transmitted. Then, when it is notified from the transmission destination computer via the data communication means 104 that backup data can be stored (step S107), the computer is determined as the transmission destination of backup data.

  That is, the terminal usage prediction unit 103 communicates with the selected computer via the data communication unit 104, and the computer that cannot communicate because the power is not turned on, or the backup data storage unit 105 stores the communication even if communication is possible. Computers that cannot receive backup data due to lack of capacity or the like are excluded from backup data transmission destination candidates.

  The terminal usage prediction unit 103 generates a backup ID that is an ID for specifying the backup data when the transmission destination of the backup data is determined, and uses the generated backup ID as the user data storage unit 101, the external storage medium Or store it on another computer or display it on the screen. That is, the terminal usage prediction unit 103 makes it possible to use the backup data by using the backup ID when the backup data becomes necessary.

  The backup ID includes a data ID for specifying backup data, a destination terminal ID for specifying a backup destination computer (a computer to which backup data is sent, in this example, a computer 100-2), and a backup. The request source terminal ID for identifying the request source computer (the computer from which the backup data is transmitted. In this example, the computer 100-1) is included.

  The backup data creation means 102 creates backup data of the data created by the user stored in the user data storage means 101 according to the confirmed backup data transmission destination computer, and sends it to the data communication means 104. . The data communication unit 104 communicates with the data communication unit 104 of the destination computer of the backup data, and transmits the backup data and the backup ID (step S108). The data communication unit 104 of the backup data transmission destination computer receives the backup data and the backup ID, and stores the received backup data and backup ID in the backup data storage unit 105 (step S109).

  Next, the case of using backup data will be described with reference to the drawings. FIG. 3 is a sequence diagram for explaining the operation when using backup data.

  When the backup ID generated when performing backup is input to the data communication unit 104 by the user (step S201), the data communication unit 104 stores the backup data based on the destination terminal ID included in the backup ID. Identify the computer that is running. Then, the data communication unit 104 transmits the backup ID and information indicating a data transmission request to the data communication unit 104 of the computer storing the backup data (step S202).

  The data communication unit 104 of the computer that has received the backup ID and the information indicating the data transmission request notifies the backup data storage unit 105 of the backup ID (step S203). The backup data storage means searches for and extracts corresponding backup data (step S204) and outputs it to the data communication means 104.

  The data communication unit 104 that has received the backup data transmits the backup data to the computer that has transmitted the backup ID and the information indicating the data transmission request (step S205). The data communication means 104 of the computer that has received the backup data passes the backup data to the backup data creation means 102 (step S206). The backup data creation unit 102 restores the original data from the received backup data (step S207) and stores it in the user data storage unit 101.

  A procedure in which the terminal usage prediction unit 103 selects a candidate for the destination of the backup data will be described. Based on the information on the operating time of the other computer collected through the data communication means 104 by predicting the operating time of the own computer in the time until the next transmission timing of the backup data to the other computer. Predict the uptime of other computers.

  As this prediction method, there is a method that is explicitly given by the user of each computer. For example, when performing backup once a day, information indicating that it is always up until the next backup, or transition to the power-off state at 5:00 pm and transition to the power-on state at 9:00 am the next day Such information is input to the terminal usage predicting means 103 by the user of each computer.

  Further, the terminal usage predicting means 103 can also predict the operation time from the past operation results for a certain period. For example, the terminal usage prediction unit 103 records the operation time for one week, and records the time when the operation starts and the time when the operation ends in a storage unit (not shown). Then, based on the recorded operation start time and operation end time, it is predicted that the operation for the next backup period will be performed in the same manner in the past for the operation having a certain probability or more.

  In addition, the computer requesting backup may collect past operation data of other computers and make predictions, or each computer may make predictions based on the operation data of its own computer and exchange prediction results with each other. May be. Further, each computer may be a peer-to-peer type, and may exchange past operation data or prediction data. A specific server collects each computer's past operation data or prediction data, and each computer collects from that server. Past operation data or prediction data may be acquired.

  The terminal usage prediction unit 103 of each computer obtains a correlation between the operation time of the own computer and the operation time of the other computer after the prediction of the operation time of the own computer and the other computer is completed. In the present embodiment, the correlation refers to the ratio of the operating time of another computer overlapping the entire operating time of the computer.

  For example, if the local computer has been operating for 8 hours from a certain time and another computer has always been operating during that 8 hours, the correlation value between the local computer and the computer is 1.0. If another computer has been operating for 4 hours during the 8 hours, the correlation value between the computer and the computer is 0.5. The terminal usage predicting means 103 defines the correlation value as the selection priority, and sends the backup data transmission destination to a predetermined number of computers having higher selection priority (high correlation value) from other computers. Select as a candidate.

  According to the present embodiment, the terminal usage prediction unit 103 exchanges information with other computers via the data communication unit 104, and the operation time overlaps with the time when the computer is predicted to be operating. Because it is configured to select other computers with a high percentage as backup data transmission destination candidates, it increases the probability that the computer storing the backup data is available when the backup data is needed be able to.

  In this embodiment, for the sake of simplicity of explanation, it has been described that the backup data transmission destination of the computer 100-1 is one computer, but there are a plurality of backup data creation means 102 of the computer 100-1. The backup data divided into two may be created, and the data communication unit 104 may transmit the backup data to a plurality of computers for storage. That is, the computer 100-1 may store the backup data in a distributed manner among a plurality of computers.

Embodiment 2. FIG.
Next, a second embodiment of the present invention will be described. In the first embodiment of the present invention, the terminal usage predicting means 103 uses the correlation between the operating time of its own computer and the operating time of the other computer in selecting the backup data transmission destination candidate.

  However, the definition of correlation in the first embodiment is not symmetric between the two computers. Therefore, in the second embodiment, a product of values obtained by further correlating each other is obtained as the simultaneous activation rate.

  For example, the simultaneous activation rate between the computers 100-1 and 100-2 is defined as the product of the correlation of the computer 100-2 viewed from the computer 100-1 and the correlation of the computer 100-1 viewed from the computer 100-2. To do. The value of the simultaneous activation rate increases in a combination of computers operating in the same time zone, and decreases in a combination of computers that are used in different time zones and do not overlap in operating time zones.

  Further, the simultaneous activation rate of the computer 100-2 when viewed from the computer 100-1 is the same as the simultaneous activation rate of the computer 100-1 when viewed from the computer 100-2.

  The terminal usage prediction means 103 calculates the simultaneous activation rate in this way, and uses the calculated simultaneous activation rate as the selection priority, and selects a predetermined number of computers among the higher selection priorities as backup data transmission destinations. Select as a candidate. That is, the terminal usage prediction unit 103 selects a computer having a high probability of operating at the same time as a backup data transmission destination candidate.

  In the present embodiment, since the terminal usage prediction unit 103 is configured to select a candidate for the backup data transmission destination using the simultaneous activation rate, the computer used by the user is more stable. Since the operation time is synchronized with the other computers so that the backup data can be accessed, the system is operated in synchronization with the entire system, so that the user does not act selfishly, the reliability of the backup, and the hardware resources of each computer Can be used for effective use of

Embodiment 3 FIG.
Next, a third embodiment of the present invention will be described. In the second embodiment, the simultaneous activation rate is used as the selection priority for selecting the transmission destination of the backup data. In the third embodiment, in addition to the simultaneous startup rate, a startup time rate that is the ratio of the time during which other computers are operating in the prediction target time is used.

  For example, when the prediction target time is 24 hours and the operating time is 8 hours, the startup time rate becomes 0.33 (rounded down to the third decimal place). For example, the terminal usage prediction unit 103 determines the selection priority by using the simultaneous activation rate and the activation time rate, with a weight α according to the following equation.

That is, the terminal usage prediction unit 103 obtains the selection priority as selection priority = simultaneous activation rate × partner terminal activation time rate ^α (where α = 0 to 1). Α is an exponent of the power of the partner terminal activation time rate. Therefore, selection priority = simultaneous activation rate × (other party activation time rate to the α power). The terminal usage predicting means 103 determines the selection priority in this way, and selects a predetermined number of computers among the higher selection priorities as backup data transmission destination candidates. Here, the value of α is given in advance by an administrator who provides the system.

  In the second embodiment, only the correlation of the operation time between the own computer and another computer is considered. When a system administrator introduces a backup server or the like that operates stably for a long time into the system, for example, simultaneous startup between a server that operates 24 hours a day and a computer that uses only 8 hours a day The rate is 0.33. In other words, the value of the simultaneous activation rate between a backup server or the like that operates stably for a long time and the computer is small. For this reason, the server may not be used as a backup destination.

  Therefore, in the present embodiment, since the terminal usage prediction unit 103 is configured so that the system administrator can set the weighting in consideration of the startup time rate, it operates for a long time according to the weighting setting of the system administrator. It is possible to increase the priority of computers and servers.

  In the first to third embodiments of the present invention, a predetermined number of computers are selected as a method for selecting backup data transmission destination candidates from among computers ordered by selection priority. You may select in order of selection priority, or you may select the number determined at random from the computer of the value of selection priority more than fixed.

  Also, instead of selecting computers at random, select a combination that maximizes the correlation between the total operating time when multiple candidate operating times are selected and combined, and the operating time of the local computer. But you can.

  Furthermore, in the first embodiment of the present specification, a case is described in which backup data is acquired when data cannot be accessed for some reason when the computer is operating. However, it can also be used to store backup data in another computer for the purpose of enabling other computers to access the data stored in the own computer when the own computer is not operating.

  In this case, in the correlation calculation for determining the selection priority, the backup data transmission destination is determined using the selection priority calculated based on the time when the own computer is not operating, not the operating time of the own computer. Select a candidate.

Embodiment 4 FIG.
Next, a fourth embodiment of the present invention will be described with reference to the drawings. FIG. 4 is a block diagram showing a configuration example of the fourth embodiment of the present invention.

  The distributed mutual backup system shown in FIG. 4 sets unit data and a price per unit time when another computer stores backup data in the own computer based on the state of the own computer and the state of the other computer. The first embodiment shown in FIG. 1 includes means 106 and budget setting means 107 for setting a budget that can be used for storing backup data based on the status of the local computer and the status of other computers. Different from the configuration.

  Since other configurations are the same as those of the first embodiment shown in FIG. 1, the same reference numerals as those in FIG.

  In the present embodiment, the price and budget are not necessarily exchangeable for actual currency. The price and budget are, for example, a virtual currency for storing backup data, or a point that can be exchanged for an actual currency at a certain exchange ratio.

  The price setting means 106 sets the price that is input by the user to the keyboard or the like provided in the computer. The budget setting means 107 sets the value input by the user on the keyboard or the like provided in the computer as the budget.

  The terminal usage prediction unit 103 receives a budget value from the budget setting unit 107. Further, the terminal usage prediction unit 103 collects price information from the terminal usage prediction unit 103 of another computer. The terminal usage prediction unit 103 refers to the price information collected from other computers, and within the range that satisfies the budget value set by the budget setting unit 107, the backup data transmission destination candidates in descending order of selection priority. Select a computer.

  A method in which the price setting unit 106 sets the price will be described. The price setting means 106 collects information indicating the status of the other computer and the status of the own computer, and takes into account various factors such as the collected information and the unit when the computer user stores backup data in the computer. Set data and price per unit time.

  One of various factors to be taken into account when the price setting means 106 sets the price is, for example, the operating time of the computer. Generally, a computer that is operating for a long time and when no other computer is operating is considered to be valuable as a storage location for backup data.

  Therefore, the price setting means 106, for example, when the basic price for storing backup data per unit time is X and the number of computers available in a unit time frame is p (including its own computer). The price at the time of storing backup data in the own computer in the time zone is set to X / p. Therefore, the sum of X / p for the entire startup time of the own computer during the backup time becomes the price when the backup data is stored in the own computer.

  By determining the price in this way, the price is high when the backup data is stored on a computer that is operating for a longer time or when no other computer is operating during the time that the backup data is stored. Can be set to

  Here, the price setting means 106 does not necessarily need to communicate with all the computers of the distributed mutual backup system to confirm the number of other computers that are operating. A computer may be extracted and estimated. Note that “close to communication network” means that the communication capacity of the communication path used for information transmission / reception is sufficient, or the communication path used for information transmission / reception supports high-speed communication, so that information can be transmitted / received at high speed. Etc.

  In addition, factors such as the reliability of the hardware resources of the computer are taken into consideration, the effect of data compression due to duplication of backup data, etc., and the server administrator stores a large amount of backup data as a service to users. The price is determined by taking into account various factors such as intentionally lowering the price.

  Next, a method in which the budget setting unit 107 sets a budget will be described. The budget setting means 107 calculates sales based on the price setting of its own computer and the amount of backup data currently stored, and sets a budget reflecting the user's intention based on the calculated sales.

  In the budget setting means 107, a budget is set based on a prediction of compensation by the user's own computer storing backup data of another computer. However, the budget setting means 107 may prepare a larger budget than the sales of its own computer using the actual currency when the virtual currency obtained as the consideration can be exchanged for the actual currency. If a budget larger than the sales of the own computer is prepared, the backup data of the own computer can be stored in another computer with higher reliability and higher reliability. The budget setting means 107 prepares a computer capable of storing a large amount of backup data as a service provider, and stores backup data contracted from another computer without preparing a budget to be deposited from the own computer to another computer. You may make a profit.

  According to the present embodiment, a computer user can arbitrarily set a price and control selection of backup data transmission destination candidates. For example, if the system administrator sets the price of a server added to the system low, it is possible to set so that backup to the server is given priority over other computers. In addition, if the virtual currency can be exchanged for the actual currency, a user who cannot always operate the computer can store backup data in another user's computer for a fee.

  Next, embodiments of the present invention will be described with reference to the drawings. FIG. 5 is an explanatory diagram showing a configuration example of an embodiment of the present invention.

  As shown in FIG. 5, the distributed mutual backup system according to the embodiment of the present invention includes 20 personal computers (computers 200-1 to 200-20) connected to each other via a LAN (Local Area Network). Further, each computer has a user data storage unit 101, a backup data creation unit 102, a terminal usage prediction unit 103, a backup data storage unit 105, a data communication unit 104, a price setting unit 106, and a budget setting unit 107 shown in FIG. including.

  The user data storage means 101 and the backup data storage means 105 are magnetic disk devices provided in each computer. The backup data creation unit 102, the terminal usage prediction unit 103, the price setting unit 106, and the budget setting unit 107 are a central processing unit that executes processing according to program control. The data communication unit 104 is a communication device provided in each computer. The price setting means 106 and the budget setting means 107 are set with a price and a budget, respectively, by a user using a keyboard provided in each computer.

  The computer 200-1 is set so that the user can store backup data in another computer once a day for a specific file group stored in the magnetic disk device. The computer 200-1 communicates with another computer to check whether the other computer is operating every hour.

  Then, the computer 200-1 communicates with other computers after the start of operation, and collects information on other computers when the computer 200-1 itself is not operating and information on past non-operations. . Then, the computer 200-1 accumulates the collected information by temporarily storing it in the magnetic disk device as log information. Based on the accumulated information for the past week, the terminal usage predicting means 103 predicts that a computer operating with a probability of 50% or more in each time zone is also operating in that time zone on the next day. .

  Next, an operation when the computer 200-1 transmits backup data to another computer will be described. The terminal usage prediction means 103 of the computer 200-1 predicts the operating time of the other computer based on information collected from the other computer.

  FIG. 6 is an explanatory diagram illustrating an example of a result of prediction of the operating time of the computer performed by the terminal usage prediction unit 103. In the example shown in FIG. 6, the vertical axis represents each computer, the horizontal axis represents a time frame in units of 3 hours, the time frame predicted to continue operating for 3 hours is indicated by diagonal lines, and the other time The frame is shown as solid as non-operational.

  FIG. 7 is an explanatory diagram showing a matrix of operating time correlations between computers based on the result of operating time prediction shown in FIG. In the example illustrated in FIG. 7, for example, the correlation of the computer 200-4 viewed from the computer 200-1 corresponds to the first row and the fourth column, and the value is 0.8. Further, the correlation of the computer 200-1 viewed from the computer 200-4 corresponds to the fourth row and the first column, and the value is 1.0.

  FIG. 8 is an explanatory diagram showing a matrix of products of the correlation matrix between the operating times of the computers shown in FIG. 7 and the transposed matrix of the matrix.

  In the matrix shown in FIG. 8, the value of the first row and the fourth column and the value of the fourth row and the first column are 0.8, which is the simultaneous activation of the computer 200-1 and the computer 200-4. Rate. Here, for example, the first row is the simultaneous activation rate of the computers 200-1 to 200-20 viewed from the computer 200-1. When this simultaneous activation rate is set as the selection priority, the terminal usage prediction unit 103 simultaneously activates from the row corresponding to each computer of the matrix illustrated in FIG. 8 as a backup data transmission destination candidate of each computer, for example. Select the three items with the highest rate values. In addition, when there are a plurality of candidates having the same value in the upper rank, they are selected at random.

  FIG. 9 is an explanatory diagram showing an example of selection of backup data transmission destination candidates based on the simultaneous activation rate by the terminal usage predicting means 103. In the example illustrated in FIG. 9, three cases having a large simultaneous activation rate value are selected from the rows corresponding to the computers in the matrix illustrated in FIG. 8, and the selected items are surrounded by a frame.

  Further, when the selection priority is determined in consideration of the terminal activation rate, it is necessary to obtain the operating time of each computer with respect to the backup interval of 24 hours. FIG. 10 is an explanatory diagram showing the terminal activation rate for each computer for 24 hours. FIG. 11 is an explanatory diagram showing the selection priority obtained by using the terminal activation rate shown in FIG. 10 and the simultaneous activation rate shown in FIG. 8 with the weight α described in the third embodiment. It is. FIG. 12 is an explanatory diagram showing a case where three candidates having a large selection priority value in each row are selected as candidates, similarly to the case where the simultaneous activation rate is set as the selection priority. In the example illustrated in FIG. 12, the terminal usage prediction unit 103 of the computer 200-1 selects the computers 200-2, 200-5, and 200-9 as backup data transmission destination candidates.

  The terminal usage predicting means 103 of the computer 200-1 thus selects the backup data transmission destination candidates, and then selects the computers 200-2, 200-5, and 200 selected as candidates via the data communication means 104. Communicate with -9 to check whether the backup data can be accepted and determine the destination of the backup data. Specifically, the terminal usage prediction unit 103 inquires of the computer selected as the backup data transmission destination candidate whether the backup data may be transmitted. When it is notified from the destination computer via the data communication means 104 that the backup data can be stored, the computer is determined as the destination of the backup data.

  The terminal usage prediction unit 103 generates a backup ID that is an ID for specifying the backup data when the transmission destination of the backup data is determined, and uses the generated backup ID as the user data storage unit 101, the external storage medium Or store it on another computer or display it on the screen. That is, the terminal usage prediction unit 103 makes it possible to use the backup data by using the backup ID when the backup data becomes necessary.

  The backup data creation means 102 creates backup data of the data created by the user stored in the user data storage means 101 according to the confirmed backup data transmission destination computer, and sends it to the data communication means 104. . The data communication unit 104 communicates with the data communication unit 104 of the backup data transmission destination computer, and transmits the backup data and the backup ID.

  The computer that has received the backup data and the backup ID stores the received backup data and backup ID in the backup data storage unit 105. In addition, the computer 200-1 stores the backup ID in the magnetic disk device.

  Next, an embodiment in the case of performing backup using a price and a budget will be described. In the description of the present embodiment, the result of the start prediction illustrated in FIG. 6 is used.

  FIG. 13 is an explanatory diagram showing an example of price setting. In the example shown in FIG. 13, in the time frames 1 to 8 in which 24 hours are divided into eight, the whole is 1, and 1 is divided by the number of computers running in the same time frame. It is set as the price per unit data.

  That is, in the time frame in which all 20 computers are operating (time frame 3 in the example shown in FIG. 13), the price per unit time frame and unit data is 1/20 = 0.05. In the time frame in which only four computers are activated (time frame 7 in the example shown in FIG. 13), the price per unit time frame and unit data is 1/4 = 0.25, so 0.25 become. In other words, the price per unit time frame and unit data when a large number of computers are operating is set low, and the price per unit time frame and unit data when a small number of computers are operating is high. Is set.

  The price set for the computer for 24 hours is obtained by adding the price per unit time frame only for the time the computer is operating for each computer. The price for each computer is shown at the right end of FIG. When the user predicts that the computer 200-1 will store backup data of about 3 unit data on average and sets a budget of 1.0 based on the price information shown at the right end of FIG. From the computer 200-1, backup data can be stored in the computers 200-2, 200-5, and 200-8, which have the same price. In addition, it is possible to make a selection such as depositing in the computer 200-9 with emphasis on the fact that the user is always activated and accessible.

  According to the present embodiment, each computer can store backup data in a distributed manner in computers that have a high probability of being available when backup data is needed.

  Further, according to the present embodiment, each computer can control the acceptance of backup data by setting a budget and a price.

  INDUSTRIAL APPLICABILITY The present invention can be applied to a system that backs up data that a user does not want to lose safely and at low cost when a plurality of users use computers in the company. Further, the present invention can be applied to a system that realizes a service in which a large number of users mutually back up data via the Internet. Furthermore, the present invention can also be applied to a system in which a service provider provides a backup service for a fee on a peer-to-peer basis via the Internet.

It is a block diagram which shows 1st Embodiment of a distributed mutual backup system. It is a sequence diagram explaining operation | movement in case a computer transmits backup data to another computer. It is a sequence diagram explaining operation | movement when using backup data. It is a block diagram which shows the structural example of 4th Embodiment. It is explanatory drawing which shows the structural example of an Example. It is explanatory drawing which shows the example of the result of prediction of the operating time of the computer which the terminal utilization prediction means performed. It is explanatory drawing which shows the matrix of the correlation of the operating time between computers based on the result of an operating time prediction. It is explanatory drawing which shows the matrix of the product of the matrix of the correlation of the working time between computers, and the transpose matrix of the matrix. It is explanatory drawing which shows the example of selection of the candidate of the transmission destination of backup data based on the simultaneous activation rate by a terminal utilization prediction means. It is explanatory drawing which shows the terminal starting rate with respect to 24 hours of each computer. It is explanatory drawing which shows the selection priority calculated | required using the terminal starting rate and the simultaneous starting rate. It is explanatory drawing which shows the case where three with a large value of selection priority are selected as a candidate. It is explanatory drawing which shows the example of price setting. 1 is a block diagram showing a system described in Patent Document 1. FIG. FIG. 10 is a block diagram illustrating a configuration of an information processing apparatus that configures a system described in Patent Literature 1.

Explanation of symbols

100-1 to 100-n Computer 101 User data storage means 102 Backup data creation means 103 Terminal usage prediction means 104 Data communication means 105 Backup data storage means 106 Price setting means 107 Budget setting means 200-1 to 200-20 Computer

Claims (13)

In a distributed mutual backup system in which a plurality of information processing devices transmit and receive backup data to each other via a communication network,
Each information processing device
Backup data creation means for creating backup data of stored data;
In order to predict the operation time of the own device and the operation time of the other information processing device, and to store the backup data based on the predicted operation time of the own device and the predicted operation time of the other information processing device And a terminal usage prediction means for selecting an information processing apparatus as a transmission destination of the backup data. The terminal usage prediction means obtains a correlation that is a ratio of overlap between the predicted operation time of the own device and the predicted operation time of another information processing device, and determines an information processing device having a high correlation of the predicted operation time as a transmission destination of backup data. The distributed mutual backup system according to claim 1 is selected. Based on the correlation between the predicted operation time of the own device and another information processing device, the terminal usage prediction means sends a predetermined number of information processing devices to the backup data transmission destination in descending order of the probability of operation at the same time. The distributed mutual backup system according to claim 2. Each information processing device
Backup data storage means for storing backup data of other information processing devices;
Price setting means for setting a price when storing backup data of another information processing apparatus in the backup data storage means;
The distributed mutual backup system according to any one of claims 1 to 3, further comprising budget setting means for setting a budget for transmitting and storing backup data in another information processing apparatus. In an information processing apparatus connected to each other via a communication network and transmitting / receiving backup data,
Backup data creation means for creating backup data of stored data;
In order to predict the operation time of the own device and the operation time of the other information processing device, and to store the backup data based on the predicted operation time of the own device and the predicted operation time of the other information processing device Terminal use predicting means for selecting an information processing apparatus as a transmission destination of the backup data. In a distributed mutual backup method in which a plurality of information processing devices connected via a communication network transmit / receive backup data to / from each other,
The terminal usage predicting means of the information processing device predicts the operation time of the own device and the operation time of the other information processing device, and based on the predicted operation time of the own device and the predicted operation time of the other information processing device. A terminal usage prediction step of selecting an information processing device as a transmission destination of the backup data in order to store backup data of the stored data;
A distributed mutual backup method, wherein the backup data creation means of the information processing apparatus includes a backup data creation step of creating backup data of stored data. The terminal usage prediction unit of the information processing device obtains a correlation that is a ratio of overlap between the predicted operation time of the own device and the predicted operation time of another information processing device in the terminal usage prediction step, and the correlation of the predicted operation time is high. The distributed mutual backup method according to claim 6, wherein the information processing apparatus is selected as a destination of backup data. The terminal usage predicting means of the information processing device is a terminal usage prediction step, based on the correlation of the predicted operating time between the own device and another information processing device, in a descending order of the probability of operating at the same time. The distributed mutual backup method according to claim 7, wherein the information processing apparatus is selected as a backup data transmission destination. A backup data storage step in which the data backup means of the information processing apparatus stores the backup data of another information processing apparatus in the backup data storage means;
A price setting step for setting a price when the price setting means of the information processing apparatus stores backup data of another information processing apparatus in the backup data storage means;
The budget setting step of the information processing apparatus includes a budget setting step of setting a budget for transmitting and storing backup data in another information processing apparatus. Distributed mutual backup method. A distributed mutual backup program installed in an information processing apparatus connected to each other via a communication network and transmitting / receiving backup data,
On the computer,
Backup data of stored data is predicted based on the predicted operation time of the own device and the predicted operation time of the other information processing device by predicting the operation time of the own device and the operation time of the other information processing device. A terminal usage prediction process for selecting an information processing device as a transmission destination of the backup data in order to store
A distributed mutual backup program that executes backup data creation processing that creates backup data of stored data. On the computer,
In the terminal usage prediction process, a correlation that is the ratio of the overlap between the predicted operation time of the own device and the predicted operation time of another information processing device is obtained, and an information processing device having a high correlation of the predicted operation time The distributed mutual backup program according to claim 10. On the computer,
In the terminal usage prediction process, based on the correlation between the predicted operation time of the own device and another information processing device, backup data is transmitted to a predetermined number of information processing devices in descending order of the probability of operating at the same time. The distributed mutual backup program according to claim 11, which is selected by an information processing device. On the computer,
Backup data storage processing for storing backup data of other information processing devices in the backup data storage means;
Price setting processing for setting a price when storing backup data of another information processing apparatus in the backup data storage means;
The distributed mutual backup program according to any one of claims 10 to 12, which causes another information processing apparatus to execute a budget setting process for setting a budget for transmitting and storing backup data.

Images