WO2013065135A1

WO2013065135A1 - Data sharing system

Info

Publication number: WO2013065135A1
Application number: PCT/JP2011/075213
Authority: WO
Inventors: 佐藤　敦; 壮一最首
Original assignee: 株式会社野村総合研究所
Priority date: 2011-11-01
Filing date: 2011-11-01
Publication date: 2013-05-10
Also published as: WO2013065545A1

Abstract

A data sharing system whereby processing such as editing can be performed on source data shared between a plurality of users, within a structure whereby the source data is segmented into a plurality of segmented data units, using a secret sharing scheme, and stored distributed in a plurality of servers, etc. A typical embodiment has: a plurality of servers (200); and a client terminal (100) that segments the source data into a n number of segmented data units, using a (k, n) threshold value secret sharing scheme, and distributes and stores each unit of data in (n−1) servers (200). The client terminal (100) segments the source data into (n−1) units of segmented data (152) and one unit of fixed segmented data (153); has a segmentation unit (110) that holds the fixed segmented data (153) in the client terminal (100), and a distribution unit (120) that sends the (n−1) units of segmented data (152) to the (n−1) servers (200); and is mounted such that the (k, n) threshold value secret sharing scheme generates the fixed segmented data (153) so as to be the same value each time, even if the content of the source data is updated.

Description

Data sharing system

The present invention relates to a technique for sharing data among a plurality of users, and in particular, based on divided data generated from a target original data by a secret sharing method and distributed and stored in a plurality of servers and the like, the other users share the original data. The present invention relates to a technology that is effective when applied to a data sharing system that shares original data by restoration.

In recent years, from the viewpoint of information security, the handling of data such as files held and processed in information processing apparatuses such as PCs (Personal Computers) used by users has been regarded as important. In particular, in addition to notebook PCs, portable terminals such as so-called smartphones and tablet PCs that are increasingly used in business need to consider the risk of information leakage due to theft or loss of these terminals themselves. There is.

On the other hand, the risk of information leakage due to loss of the terminal due to the so-called thin client that stores data including important data in the terminal in an external data center or server where security measures are taken It is conceivable to reduce. At this time, the important data is not stored in an external server or the like as it is, but for example, the so-called secret sharing technique described in Non-Patent Document 1 or the like is used, and the important data alone is meaningless. It has also been proposed to divide into non-important data (important data cannot be restored / inferred) and store these non-important data in a plurality of external servers. Thereby, for example, the risk of information leakage can be reduced even in the case of storage in a virtual data center or virtual server in a cloud computing environment.

In addition, when important data is divided into a plurality of divided data by a secret sharing technique, even if a part of the divided data is lost, the original important data can be restored if a predetermined number or more of divided data can be collected. Can also improve the availability of data. For example, when important data is divided into n pieces of divided data by so-called (k, n) threshold type secret sharing, the important data can be restored if k or more pieces of divided data can be collected. In other words, it is possible to withstand the loss of up to (n−k) pieces of divided data. Utilizing such high availability, it is also considered that the divided data is distributed and stored in a plurality of remote locations to be used as a backup considering disaster countermeasures.

As described above, in a mechanism in which a plurality of divided data obtained by dividing the original data by the secret sharing technique is distributed and stored in a plurality of servers, etc., each user individually manages the divided data. It leads to a decrease in convenience. For example, it is necessary for the user to know the number of divided data generated from the original data, each file name, on which server each divided data is stored, and to restore the original data. It is difficult to force the user to operate by specifying a server or the like in which each piece of divided data is stored.

Therefore, normally, the system performs processing such as secret sharing processing (processing for dividing and restoring original data) that is performed when the original data is stored and referenced, distributed storage of generated divided data, and collection of divided data. It is automatically performed to conceal parts related to these processes from the user.

For example, in an information processing apparatus that performs secret sharing processing for dividing user original data into a plurality of divided data, such as a client terminal of each user or a specific management server such as a file server, which server or the like Management information including the location information indicating whether or not it has been stored (hereinafter, sometimes referred to as “distributed management information”) is stored. The distributed management information is associated with, for example, a dummy file representing the original data, and the user performs operations such as reference and editing on the dummy file.

Upon receiving an operation request for the original data from the user via the dummy file, the information processing apparatus refers to the distributed management information to identify the server where the necessary divided data is stored, and directly Access the server etc., collect the necessary divided data and restore the original data. As a result, the user can access the original data distributed and stored as a plurality of pieces of divided data by using an interface similar to a normal file operation.

For example, Japanese Patent Laid-Open No. 2007-213405 (Patent Document 1) stores tally folders A, B,... For storing tally files, a restoration destination folder for storing restoration files, and a tally object file by an information management computer. A tally object folder, a tally engine folder containing a restoration engine program and a division engine program, and a tally parameter including information on a decoding boundary, which is a range that can be read by the tally application, are set as tally object files A, B,. The tally file name / storage location and the object information of the restoration destination folder are stored in, the tally file is collected directly based on the tally file storage location and the decoding boundary, the restoration file is generated, and the restoration file is stored and opened. By the secret sharing method Distributed information file management means for restoring efficiently locate and original data to prevent file is described.

JP 2007-213405 A

In the prior art, the original data is divided into a plurality of divided data by secret sharing, and these are distributed and stored, so that the original data can be securely stored while increasing the availability. At this time, for example, the original data distributed and stored in the server or the like as a plurality of divided data is shared with other users so that the normal file is placed on the file server or the like instead of on the local client terminal. There may naturally be a desire to share with users.

At this time, if only reference sharing is performed without editing or updating the original data, it can be realized relatively easily in the prior art as in the case of normal file sharing. For example, the distribution management information and dummy files generated when the original data is divided by the secret sharing method (for example, the tally object file in the example described in Patent Document 1) and the creator of the original data and the sharer Sync online with or offline. Accordingly, the sharer can collect necessary divided data based on the synchronized distributed management information, and can restore and refer to the original data.

However, with the prior art, it is difficult to realize sharing with editing and updating of the original data while maintaining consistency. For example, when sharing distributed management information etc. by synchronizing multiple users as described above, the sharer restored the original data based on the synchronized distributed management information and edited this If secret sharing and distributed storage are performed again later, the newly generated or updated divided data and distributed management information will be different from those before editing. Further, depending on the system configuration, the location of a server or the like where each divided data is stored may be different from that before editing.

That is, the distributed management information held by each sharer specifies different versions of divided data for the same original data. Therefore, for example, a sharer other than the sharer who performed the edit cannot grasp that the edit has been performed, and the original data before the edit even if the original data is restored based on the distributed management information that the sharer has. Is restored, and the consistency of the original data among the sharers cannot be maintained. In order to solve this problem, it is necessary to perform processing such as re-synchronizing the distributed management information corresponding to the original data after editing between the sharers.

Therefore, an object of the present invention is to divide original data into a plurality of divided data by a secret sharing method and distribute the original data in a plurality of servers etc. to securely store the original data among a plurality of users. An object of the present invention is to provide a data sharing system that enables processing of editing and updating original data without requiring resynchronization processing for distributed management information and the like when sharing data. The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

Of the inventions disclosed in this application, the outline of typical ones will be briefly described as follows.

A data sharing system according to a typical embodiment of the present invention is connected to a plurality of servers having storage devices and each of the servers via a network, and n original data is obtained by a (k, n) threshold secret sharing method. The data sharing system includes an information processing apparatus that divides each of the divided data into the divided data and distributes and stores the divided data in the storage devices of the n servers. The data sharing system has the following characteristics.

That is, the information processing apparatus divides the original data into (n−1) first divided data and one second divided data by the (k, n) threshold secret sharing method. And (n-1) pieces of the first divided data are transmitted to the (n-1) different servers, each of which holds the second divided data on the information processing apparatus. And the (k, n) threshold secret sharing method generates the second divided data so as to have the same value every time even when the contents of the original data are updated. Has been implemented. In addition, the server includes a distributed storage unit that stores the first divided data transmitted from the information processing apparatus in the storage device.

Among the inventions disclosed in the present application, effects obtained by typical ones will be briefly described as follows.

According to a typical embodiment of the present invention, in a mechanism for securely storing original data by dividing the original data into a plurality of divided data by the secret sharing method and distributing and storing them in a plurality of servers or the like. When the original data is shared among a plurality of users, it becomes possible to perform editing and update processing on the shared original data without requiring resynchronization processing for the distributed management information and the like.

It is the figure which showed the outline | summary about the structural example of the data sharing system which is one embodiment of this invention. It is the figure which showed the outline | summary about the example in the case of performing the sharing accompanying the edit and update with respect to the original data in one embodiment of this invention. It is the figure which showed the outline | summary about the example of the flow of a process at the time of preserve | saving original data in the client terminal in one embodiment of this invention. It is the figure which showed the outline | summary about the example of the flow of a process at the time of decompress | restoring original data in the client terminal in one embodiment of this invention. It is the figure which showed the outline | summary about the example in the case of performing the sharing accompanying the edit and update with respect to the original data in a prior art.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted. In the following, in order to make the features of the present invention easier to understand, the description will be made in comparison with the prior art.

<Overview>
As described above, in the conventional technique, the original data is divided into a plurality of divided data by the secret sharing method, and these are distributed and stored, so that the original data can be securely stored while increasing the availability. Here, when sharing original data that is distributed and stored in a server or the like as a plurality of divided data with other users, if the reference data is not edited or updated, only normal file sharing As in the case of, it can be realized relatively easily. For example, the distribution management information and dummy files generated when the original data is divided by the secret sharing method are synchronized online or offline between the creator of the original data and the sharer. Accordingly, the sharer can collect necessary divided data based on the synchronized distributed management information, and can restore and refer to the original data.

However, with the conventional technology, it is difficult to realize sharing with editing and updating of the original data while maintaining consistency. FIG. 5 is a diagram showing an outline of an example in the case of sharing with editing and updating of original data in the conventional technique. In the example of FIG. 5, first, in the client terminal A (100a), when the user stores the original data 150a (version 1), for example, n pieces of original data 150a (see FIG. The data is divided into four pieces of divided data 152a (version 1), and these are distributed and stored in n servers 200. Here, for convenience of explanation, a plurality of divided data 152a is illustrated as being stored together in the server 200, but in reality, each divided data 152a is distributed and stored on n different servers 200. And

At this time, the client terminal A (100a) specifies each divided data 152a corresponding to the original data 150a, and generates distributed management information 151a (version 1) including information relating to the location of each divided data 152a. . The distributed management information 151a may be associated with a dummy file (for example, a so-called shortcut or alias for the original data 150a) corresponding to the original data 150a.

Thereafter, the distributed management information 151a is synchronized with another client terminal B (100b). For example, the distributed management information 151a can be transmitted to the client terminal B (100b) via the network 300 and synchronized online, or can be manually transferred to the user of the client terminal B (100b) and synchronized offline. .

In the client terminal B (100b) that has synchronized the distributed management information 151a, the client terminal B (100b) refers to the distributed management information 151a based on a request from the user or the like, and is k or more from the server 200 in which each divided data 152a is stored. The divided data 152a is acquired and collected. Further, the original data 150a is restored by the (k, n) threshold secret sharing method based on the collected k or more pieces of divided data 152a.

Thereafter, the user of the client terminal B (100b) edits or updates the original data 150a with a predetermined application or the like to obtain the edited original data 150b (version 2). Similarly, when storing the original data 150b, the original data 150b is divided into n pieces (4 pieces in the example of FIG. 5) of divided data 152b (version 2) by, for example, the (k, n) threshold secret sharing method. These are distributed and stored in n servers 200. The n servers 200 here may be partly or entirely different from the n servers 200 in which the divided data 152a (version 1) is stored. In addition, the client terminal B (100b) specifies each divided data 152b corresponding to the original data 150b, and generates distributed management information 151b (version 2) including information related to the location of the divided data 152b.

In this state, in the client terminal A (100a), what is specified by the distributed management information 151a (version 1) is the divided data 152a (version 1), not the divided data 152b (version 2). Therefore, unlike the client terminal B (100b), the latest original data 150b (version 2) cannot be restored. In order to make this consistent, it is necessary to synchronize the distributed management information 151b generated at the client terminal B (100b) with the client terminal A (100a) again.

Therefore, in the data sharing system according to an embodiment of the present invention, when the original data 150 is divided into a plurality of divided data 152 by the (k, n) threshold secret sharing method, the original data 150 is edited or updated. Even when the content is changed after being performed, each divided data 152 is generated so that one divided data always has the same value. The divided data generated so as to have the same value each time (hereinafter sometimes referred to as “fixed divided data”) is not stored in the server 200 in a distributed manner, but the original data 150 is divided by the secret sharing process. Each of the client terminals 100 divided into data 152 is held and shared among a plurality of users.

In each client terminal 100, by using this fixed division data instead of the distributed management information 151, each server does not require the resynchronization processing of the distributed management information 151 accompanying the editing or updating of the original data 150. The corresponding divided data 152 (that is, the latest version of the divided data 152) is collected from 200, and the latest version of the original data 150 can be restored.

FIG. 2 is a diagram showing an outline of an example in the case of sharing with editing or updating of original data in the data sharing system according to an embodiment of the present invention. In the example of FIG. 2, as in the example of FIG. 5, first, when the user stores the original data 150a (version 1) in the client terminal A (100a), for example, by the (k, n) threshold secret sharing method. The original data 150a is divided into a total of n pieces of divided data consisting of (n−1) pieces (three pieces in the example of FIG. 2) pieces of divided data 152a (version 1) and one piece of fixed divided data 153. As described above, the fixed divided data 153 is divided data generated so as to have the same value every time even when the content is changed by editing or updating the original data 150a.

In the example of FIG. 2, (n-1) pieces of divided data 152a among the generated n pieces of divided data are distributed and stored in a plurality of servers 200. As in the example of FIG. 5, for convenience of explanation, a plurality of pieces of divided data 152a are shown to be stored together in the server 200, but actually, each piece of divided data 152a is different from each other (n−1) pieces. Are distributed and stored on the server 200. At this time, the client terminal A (100a) holds the fixed divided data 153. Note that the fixed division data 153 may be associated with a dummy file corresponding to the original data 150a (for example, a so-called shortcut or alias for the original data 150a).

Thereafter, the fixed division data 153 is synchronized with the other client terminal B (100b). For example, the fixed division data 153 can be transmitted to the client terminal B (100b) via the network 300 and synchronized online, or can be manually passed to the user of the client terminal B (100b) and synchronized offline. . If the client terminal B (100b) already has the fixed divided data 153 due to synchronization in the past or the like, the resynchronization process is unnecessary.

In the client terminal B (100b) that has synchronized the fixed division data 153, based on the request from the user, etc., the (k−1) or more division data 152a corresponding to each server 200 based on the fixed division data 153. Acquire and collect Further, based on the collected (k−1) or more pieces of divided data 152a and one piece of fixed divided data 153 held by itself, a total of k or more pieces of divided data, the (k, n) threshold secret sharing method Thus, the original data 150a is restored.

There are several methods for collecting the corresponding divided data 152a based on the fixed divided data 153, and these can be applied as appropriate. For example, when the original data 150a is divided into the fixed divided data 153 and the divided data 152a by the secret sharing method in the client terminal A (100a), information such as a file ID for identifying the original data 150 is changed to the fixed divided data 153 and each It is added to the header of the divided data 152a. Thereby, the divided data 152a including the file ID information corresponding to the file ID information included in the header or the like of the fixed divided data 153 in the header or the like can be identified.

Thereafter, the user of the client terminal B (100b) edits or updates the original data 150a with a predetermined application or the like to obtain the edited original data 150b (version 2). Similarly, when the original data 150b is stored, (n−1) (three in the example of FIG. 2) divided data 152b (version 2) of the original data 150b is obtained by, for example, the (k, n) threshold secret sharing method. ) And one fixed divided data 153 to be divided into a total of n divided data. Note that the value of the fixed division data 153 is unchanged from the value of the fixed division data 153 of the version before editing as described above.

In the example of FIG. 2, (n−1) pieces of divided data 152b among the generated n pieces of divided data are distributed and stored in (n−1) servers 200. Here, the (n−1) servers 200 may be partially or entirely different from the (n−1) servers 200 in which the divided data 152a (version 1) is stored. Further, the client terminal B (100b) holds the fixed divided data 153. At this time, the fixed division data 153 may be associated with a dummy file or the like corresponding to the original data 150b.

In this state, unlike the example of FIG. 5, in the client terminal A (100a), the content of the fixed division data 153 is the same as that in the client terminal B (100b). The latest latest divided data 152b (version 2) corresponding to the stored fixed divided data 153 is identified and collected based on the held fixed divided data 153 without the need for resynchronization processing, and the latest original data 150b (version 2) is collected. It is possible to restore. For this purpose, it is necessary to delete the old version of the divided data 152a in each server 200. That is, generation management cannot be performed, and it is necessary that each server 200 always has the latest version of the divided data 152.

<System configuration>
FIG. 1 is a diagram showing an outline of a configuration example of a data sharing system according to an embodiment of the present invention. In the data sharing system 1, a plurality of client terminals 100 sharing the original data 150 (two in the example of FIG. 1, client terminal A (100 a) and client terminal B (100 b)) and a plurality of servers 200 are connected to the Internet or the like. Are connected to each other via the network 300 and can communicate with each other.

The client terminal 100 is an information processing apparatus such as a PC or a portable terminal that is used by the user to create, edit, and store original data 150 including important data. ) The division processing unit 110 (110a, b), the distributed processing unit 120 (120a, b), the restoration processing unit 130 (130a, b), and the interface unit 140 (140a, b) implemented by the software program that operates on the above. Etc. have each part. Further, the above-described fixed divided data 153 generated from the original data 150 by the division processing unit 110 is held on a storage device such as an HDD (Hard Disk Disk Drive) (not shown).

For example, the division processing unit 110 performs, for example, (k, n) threshold secret sharing (k ≦ n) on the original data 150 instructed to be securely stored by the user via the interface unit 140 described later according to a predetermined procedure. Thus, the data is divided into (n−1) pieces of divided data 152 distributedly stored in each server 200 and one piece of fixed divided data 153 held on the client terminal 100. Note that the secret sharing algorithm is not particularly limited, and a known method can be used. However, as described above, even if editing or updating is performed on certain original data 150, fixed divided data that has the same value every time. It is assumed that it has a function of generating one 153.

In the generally known secret sharing technique, the function of generating the fixed division data 153 as described above can be realized relatively easily. For example, in the case of the (2, 3) threshold secret sharing method using a polynomial, the slope is such that the secret information S represented by the intercept always passes a specific coordinate in the straight line represented by the equation y = ax + S. Set the value of a. In this straight line, the divided data obtained based on the specific coordinates is set as the fixed divided data 153, and the other two divided data 152 are determined based on the coordinates of the other two points on the randomly selected straight line. be able to.

For example, the distributed processing unit 120 transmits (n−1) pieces of divided data 152 generated from the original data 150 by the division processing unit 110 to each server 200 according to a predetermined condition based on the content of setting information (not shown). And distributed storage. The setting information includes, for example, access information (IP address, host name, etc.) for each server 200 serving as a distributed storage destination, and when there are more than (n-1) servers 200, (n-1) pieces of information are set. Information such as criteria and conditions for selecting the server 200 (for example, priority of the server 200, an ordered list, a rotation method, etc.) can be set in advance by a file, a registry, or the like.

The distributed processing unit 120 restores the original data 150 from each server 200 based on a request from the restoration processing unit 130 when restoring the original data 150 by the restoration processing unit 130 described later. The divided data 152 ((k−1) ≦ m ≦ (n−1)) is collected and transferred to the restoration processing unit 130. As a method of collecting the m pieces of divided data 152, for example, as described above, each server 200 has the corresponding divided data 152 based on the fixed divided data 153 corresponding to the original data 150. It is collected from each server 200 having this by inquiring whether or not it exists.

Note that when one or more of the (n−1) pieces of divided data 152 cannot be stored in each server 200 due to a failure of the server 200 or the like when the divided data 152 is distributed and stored, If (k−1) or more cannot be collected from time to time, an error may be returned to the user. Further, when transmitting / receiving the divided data 152 to / from each server 200, the client terminal 100 and each server 200 perform transmission / reception after performing predetermined encryption on the divided data 152, thereby causing information leakage. This risk may be further reduced.

For example, the restoration processing unit 130 has the number of (k−1) or more necessary for restoring the original data 150 instructed to be used for reference or editing by the user via the interface unit 140. The division data 152 is requested and acquired from the distributed processing unit 120. Furthermore, according to a predetermined procedure, for example, (k, n) is obtained from a total of k pieces or more of the obtained (k−1) or more pieces of divided data 152 and one fixed piece of divided data 153 held by itself. ) Restore the original data 150 by the threshold secret sharing method.

The interface unit 140 has a user interface such as a screen display in the client terminal 100 and an input / output function such as data transmission / reception. The user can use the functions of the client terminal 100 by using, for example, a file management screen of a general OS.

For example, on the file management screen, the user moves important data to a specific folder or the like by a simple operation such as drag and drop. With this as a trigger, the division processing unit 110 and the distribution processing unit 120 automatically divide the important data into (n-1) pieces of divided data 152 and one piece of fixed divided data 153 as original data 150. The divided data 152 and the fixed divided data 153 can be distributed and stored in each server 200 or the like without making the user aware of it. At this time, the original data 150 is deleted from the client terminal 100, but the fixed divided data 153 can be specified corresponding to the original data 150, for example, so as not to make the user aware of it on the file management screen. A dummy file or the like (not shown) may be created and left.

Further, for example, the user performs operations such as reference and editing on the original data 150 by performing operations on the dummy file of the original data 150 managed in a specific folder on the file management screen. be able to. That is, using the operation on the dummy file or the like as a trigger, the distributed processing unit 120 and the restoration processing unit 130 automatically set m ((k−) from each server 200 based on the fixed division data 153 specified by the dummy file or the like. 1) The corresponding division data 152 corresponding to ≦ m ≦ (n−1)) can be collected, and the original data 150 can be restored from these and the fixed division data 153 to be made available to the user.

As will be described later, for example, the fixed divided data 153 generated from the original data 150 at the client terminal A (100a) is synchronized with the other client terminal 100B (100b) online or offline, and the client terminal B ( 100b), a dummy file or the like for the fixed divided data 153 is newly generated. Thereby, on the client terminal B (100b), it is possible to perform operations such as reference and editing on the corresponding original data 150 by the same operation via the dummy file or the like, and the client terminal A (100a) and the original The data 150 can be shared. Instead of using a dummy file or the like, the fixed divided data 153 may be used as it is, and the original data 150 may be restored and made available to the user in response to an operation instruction from the user.

In the example of FIG. 1, each client terminal 100 individually performs processing such as division and restoration of the original data 150 by the secret sharing method, distributed storage in each server 200, etc. This process may be executed collectively on a specific server such as a file server that stores the original data 150.

The server 200 is an information processing apparatus having a storage device such as an HDD (not shown) that can store the divided data 152 transmitted from the client terminal 100, and includes, for example, a file server or a storage server. Moreover, the data center which has these information processing apparatuses may be sufficient. Further, it may be a virtual server or a virtual data center by a cloud computing service.

The server 200 includes, for example, a distributed storage unit 210 that is implemented by a software program that runs on an OS (not shown). The distributed storage unit 210 stores the divided data 152 transmitted from the client terminal 100 in the storage device. Further, in response to the inquiry about the divided data 152 from the client terminal 100, whether or not the divided data 152 corresponding to the designated fixed divided data 153 is stored is searched, and the corresponding divided data 152 is stored. In this case, the divided data 152 is returned to the client terminal 100. At this time, the divided data 152 may be deleted from the storage device (the old version of the divided data 152 is not left).

Here, as described above, there are several methods for determining whether or not each stored divided data 152 corresponds to the designated fixed divided data 153. It can be applied as appropriate. For example, when the original data 150a is divided into the fixed divided data 153 and the divided data 152a by the secret sharing method at the client terminal A (100a), the original data 150 added to the header of the fixed divided data 153 and each divided data 152a, etc. The divided data 152 having the same file ID as the fixed divided data 153 can be searched based on information such as the file ID for identifying the file.

<Process flow>
FIG. 3 is a diagram showing an overview of an example of the flow of processing when the original data 150 is stored in the client terminal 100. For example, when the user instructs storage of the original data 150 by an operation via the interface unit 140a on the client terminal A (100a), first, the original data 150 is divided into a plurality of divided data by the secret sharing method by the division processing unit 110a. The data is divided into 152 and one fixed divided data 153 (S01). For example, the data is divided into (n−1) pieces of divided data 152 and one piece of fixed divided data 153 by the (k, n) threshold secret sharing method.

Next, the distributed processing unit 120a transmits (n-1) pieces of divided data 152 to different (n-1) servers 200 determined based on a predetermined rule (S02). FIG. 3 shows an example in which the divided data 152 is transmitted to each of the server A (200a) and the server B (200b). Each server 200 that receives the divided data 152 stores the received divided data 152 in the storage device by the distributed storage unit 210 (S03), and returns the processing result to the client terminal A (100a).

In the client terminal A (100a), the distributed processing unit 120a determines whether all (n-1) pieces of divided data 152 have been normally stored in the server 200 (S04). Here, if any of the (n−1) divided data 152 could not be stored normally, an error may be notified to the user via the interface unit 140a. At this time, the series of processes described above may be rolled back. Further, even when there is divided data 152 that has not been normally stored, if the storage of (k-1) or more divided data 152 has been completed normally, the original data 150 can be restored, and an error will occur. You may not make it.

When the distributed storage to each server 200 is completed normally, the division processing unit 110a stores the fixed division data 153 generated in step S01 in a storage device or the like (S05). At this time, a dummy file that can specify the fixed divided data 153 corresponding to the original data 150 may be generated. The divided data 152 distributed and stored in each server 200 may be deleted from the storage device of the client terminal A (100a). Moreover, you may make it synchronize (transmit) the fixed division | segmentation data 153 with respect to the other client terminal 100 which shares the original data 150 as needed. At this time, the corresponding dummy file or the like may also be synchronized.

On the other hand, FIG. 4 is a diagram showing an outline of an example of a flow of processing when the original data 150 is restored in the client terminal 100. For example, on the client terminal B (100b) that has received the synchronization of the fixed division data 153 from the client terminal A (100a), the user references (edits or updates) the original data 150 by operating the dummy file via the interface unit 140b. First, the restoration processing unit 130b acquires the fixed divided data 153 corresponding to the target original data 150 from the storage device (S11). For example, the corresponding fixed divided data 153 is specified and acquired based on information such as a dummy file operated by the user.

Next, the distributed processing unit 120b inquires of each server 200 whether or not the divided data 152 corresponding to the fixed divided data 153 is held (S12). Here, for example, an inquiry message is broadcast to each server 200 (or multicasted to servers 200 within a predetermined range). FIG. 4 shows an example in which an inquiry message is broadcast (or multicast) to each server 200 including the server A (200a) and the server B (200b).

In the server 200 that has received the inquiry message, the distributed storage unit 210 searches whether the divided data 152 corresponding to the fixed divided data 153 is held (S13). For example, as described above, the file ID is the same as that of the fixed divided data 153 based on information such as the file ID for identifying the original data 150 added to the fixed divided data 153 and the header of each divided data 152. The divided data 152 is searched.

For example, when the corresponding divided data 152 is included as in the server A (200a), it is transmitted to the client terminal B (100b) (S14). On the other hand, when the corresponding divided data 152 is not included as in the server B (200b), the fact may be transmitted to the client terminal B (100b).

In the client terminal B (100b), the distributed processing unit 120b determines whether or not the number m of the divided data 152 that can be collected is equal to or more than (k−1) necessary for restoring the original data 150. (S15). Here, if (k−1) or more pieces of divided data 152 cannot be collected, an error may be notified to the user via the interface unit 140b.

When (k-1) or more pieces of divided data 152 can be collected, the restoration processing unit 130b uses the collected (k-1) or more pieces of divided data 152 and the fixed divided data 153 acquired in step S11. The original data 150 is restored by the (k, n) threshold secret sharing method (S16). The original data 150 restored here is the latest version as shown in FIG. The original data 150 is presented to the user via the interface unit 140b, and the user can perform processing such as reference, editing, and updating.

When storing the original data 150 after being edited or updated, the original data 150 is again divided by secret sharing and the divided data 152 is sent to each server 200 by the series of processes shown in FIG. Distributed storage is performed. As a result, the user of the client terminal A (100a) also restores the latest version of the original data 150 based on the fixed division data 153 held by the client terminal A (100a), and refers to, edits, etc. The original data 150 can be shared between the client terminal A (100a) and the client terminal B (100b).

As described above, according to the data sharing system 1 according to an embodiment of the present invention, when the original data 150 is divided into the plurality of divided data 152 by the (k, n) threshold secret sharing method, Even if the data 150 is edited or updated and the content changes, the fixed division data 153 is generated so that it always has the same value. The fixed divided data 153 is not stored in the server 200 in a distributed manner but is held on each client terminal 100 obtained by dividing the original data 150 into the divided data 152 by the secret sharing process and is shared among a plurality of users. . As a result, each client terminal 100 collects the corresponding divided data 152 from each server 200 by using the fixed divided data 153 without requiring re-synchronization processing associated with editing or updating of the original data 150. Thus, the latest version of the original data 150 can be restored, and sharing of the original data 150 can be realized.

As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

The present invention relates to a data sharing system in which original data is shared by other users restoring original data based on divided data generated from a target original data by a secret sharing method and distributed and stored in a plurality of servers. Is available.

1 ... Data sharing system,
100 (100a, b) ... Client terminal, 110 (110a, b) ... Division processing unit, 120 (120a, b) ... Distributed processing unit, 130 (130a, b) ... Restore processing unit, 140 (140a, b) ... Interface unit, 150 (150a, b) ... original data, 151 (151a, b) ... distributed management information, 152 (152a, b) ... divided data, 153 ... fixed divided data,
200 ... server, 210 ... distributed storage unit,
300: Network.

Claims

A plurality of servers having storage devices;
The server is connected to each server via a network, the original data is divided into n pieces of divided data by a (k, n) threshold secret sharing method, and each piece of the divided data is respectively stored in the storage devices of the n pieces of servers. A data sharing system having an information processing apparatus for distributed storage,
The information processing apparatus includes:
The original data is divided into (n−1) first divided data and one second divided data by the (k, n) threshold secret sharing method, and the second divided data is divided into A division processing unit held on the information processing apparatus;
A distributed processing unit that transmits (n−1) pieces of the first divided data to different (n−1) pieces of the servers,
The (k, n) threshold secret sharing method is implemented to generate the second divided data so as to have the same value every time even when the contents of the original data are updated.
The server
A data sharing system, comprising: a distributed storage unit that stores the first divided data transmitted from the information processing apparatus in the storage device.
The data sharing system according to claim 1,
The information processing apparatus includes:
And a restoration processing unit that restores the original data from the (k−1) or more first divided data and the second divided data by the (k, n) threshold secret sharing method. ,
The distributed processing unit of the information processing apparatus includes:
The second divided data corresponding to the original data to be restored is specified, and each server is inquired whether the first divided data corresponding to the second divided data is held. Broadcast a message, collect (k−1) or more of the first divided data and pass it to the restoration processing unit,
The distributed storage unit of each server is
It is searched whether or not the first divided data corresponding to the second divided data designated in the message is stored in its own storage device, and if it is stored, the corresponding first divided data is stored. Is transmitted to the information processing apparatus.
The data sharing system according to claim 1 or 2,
The information processing apparatus includes:
When dividing the original data into a plurality of pieces of divided data by the division processing unit, a dummy file that can specify the second divided data is generated and held corresponding to the original data. Characteristic data sharing system.
The data sharing system according to any one of claims 1 to 3,
The information processing apparatus includes:
The second divided data generated when the original data is divided into a plurality of the divided data by the division processing unit is transmitted to another information processing apparatus that does not have the second divided data. A data sharing system characterized by that.