OBJECT MANAGEMENT METHOD, COMPUTER SYSTEM, AND OBJECT MANAGEMENT PROGRAM

The method secures custom metadata in object storage by storing it as an anonymized second object, addressing the vulnerability of existing metadata protection techniques.

JP7579318B2Active Publication Date: 2024-11-07HITACHI VANTARA LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
JP2022205140
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2022-12-22
Publication Date
2024-11-07
Estimated Expiration
2042-12-22

AI Technical Summary

Technical Problem

Existing techniques for protecting metadata in object storage systems fail to secure custom metadata added to actual data, leaving it vulnerable.

Method used

An object management method that involves storing actual data with custom metadata in a first object and anonymizing the custom metadata to create a second object, which is stored separately, ensuring security protection of the custom metadata.

Benefits of technology

This approach effectively secures custom metadata by anonymizing it, maintaining data integrity and confidentiality within object storage systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007579318000001
    Figure 0007579318000001
  • Figure 0007579318000002
    Figure 0007579318000002
  • Figure 0007579318000003
    Figure 0007579318000003
Patent Text Reader

Abstract

To realize information protection of custom metadata of real data stored in an object storage.SOLUTION: An object management method executed by a computer system includes responding to a storage request to store real data to which custom metadata, which is metadata customizable by a user, is added, in a storage area, to store a first object including the real data and the custom metadata in the storage area. Further, the method includes creating anonymized custom metadata obtained by anonymizing the custom metadata. Further, the method includes storing a second object including the anonymized custom metadata in the storage area.SELECTED DRAWING: Figure 7
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an object management method, a computer system, and an object management program. [Background technology]

[0002] In recent years, object storage has been attracting attention as a storage suitable for data lake infrastructure. In object storage, actual data, metadata, and custom metadata are stored as a single object.

[0003] Metadata is data that is assigned to real data by a system when the real data is stored in object storage. Custom metadata is metadata that is assigned to real data by a user and is customizable by the user and is used for searching and analyzing the real data. Objects stored in object storage are widely used for various analyses, training data for machine learning, etc., by utilizing metadata and custom metadata.

[0004] Under such circumstances, a technique for concealing metadata and setting access restrictions for each metadata has been devised as a technique for protecting the confidentiality of metadata (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2019-082887 A Summary of the Invention [Problem to be solved by the invention]

[0006] However, while the technology disclosed in the above-mentioned Patent Document 1 realizes confidentiality protection for metadata, it cannot realize confidentiality protection for custom metadata added to actual data.

[0007] The present invention has been made in consideration of the above-mentioned background, and has an object to realize confidentiality protection of custom metadata added to actual data stored in object storage. [Means for solving the problem]

[0008] In one aspect of the present invention, there is provided an object management method executed by a computer system having a storage area in which objects are stored, the computer system having a processor cooperating with a memory, and characterized in that the processor, in response to a storage request to store actual data to which custom metadata, which is metadata customizable by a user, is assigned in the storage area, stores a first object including the actual data and the custom metadata in the storage area, creates anonymized custom metadata by anonymizing the custom metadata, and stores a second object including the anonymized custom metadata in the storage area. Effect of the Invention

[0009] According to one aspect of the present application, it is possible to realize confidentiality protection of custom metadata added to actual data stored in object storage. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing the configuration of an information processing system according to an embodiment. [Diagram 2] An illustration of an object storage region. [Diagram 3] A diagram showing a region map. [Figure 4] FIG. 11 is a sequence diagram showing a process when an object is stored according to the embodiment. [Diagram 5] FIG. 2 is a diagram showing a configuration of metadata according to the embodiment. [Figure 6] Diagram showing metadata stored in a region. [Figure 7] FIG. 11 is a sequence diagram showing a custom metadata anonymization process according to the embodiment. [Figure 8] FIG. 4 is a sequence diagram showing a data access process according to the embodiment. [Figure 9A] FIG. 11 is a sequence diagram showing an object update process (when versioning is disabled) according to the embodiment. [Figure 9B] FIG. 11 is a sequence diagram showing an object update process (when versioning is enabled) according to the embodiment. [Figure 10] FIG. 4 is a sequence diagram showing an object deletion process according to the embodiment. [Figure 11] FIG. 11 is a sequence diagram showing a process when data is corrupted according to the embodiment. [Figure 12] FIG. [Figure 13] A hardware diagram of the computer that realizes each node. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, an embodiment of an object management method, a computer system, and an object management program according to the disclosure of the present application will be described with reference to the drawings. The embodiment, including the drawings, is an example for explaining the present application. In the embodiment, appropriate omissions and simplifications are made to clarify the explanation. Unless otherwise specified, the components of the embodiment may be singular or plural. Furthermore, a form in which one embodiment is combined with another embodiment is also included in the embodiment according to the present application.

[0012] The same or similar components are given the same reference numerals, and in the following embodiments and examples, the description may be omitted or only the differences may be described. In addition, when there are multiple identical or similar components, the same reference numerals may be given different subscripts to describe them. In addition, when there is no need to distinguish between these multiple components, the subscripts may be omitted. The number of each component may be singular or plural unless otherwise specified.

[0013] In the embodiments, the processing performed by executing a program may be described. A computer performs processing defined in a program using a processor (e.g., a CPU (Central Processing Unit), a GPU (Graphics Processing Unit)) while using memory in a main storage device. Therefore, the processor may be the subject of processing performed by executing a program. The processor executes the program to realize a functional unit that performs processing.

[0014] Similarly, the subject of the processing performed by executing the program may be a controller, device, system, computer, or node having a processor. The subject of the processing performed by executing the program may be a computing unit, and may include a dedicated circuit that performs specific processing. The dedicated circuit is, for example, an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit).

[0015] The program may be installed in the computer from a program source. The program source may be, for example, a program distribution server or a non-transitory storage medium readable by the computer. When the program source is a program distribution server, the program distribution server may include a processor and a storage resource (storage) for storing the program to be distributed, and the processor of the program distribution server may distribute the program to be distributed to other computers. In addition, in the embodiment, two or more programs may be realized as one program, and one program may be realized as two or more programs.

[0016] In the following embodiments, a DB (Data Base) may be used as a means for storing information, but the means for storing information is not limited to a DB.

[0017] [Embodiment] (Configuration of Information Processing System S) 1 is a diagram showing the configuration of an information processing system S according to an embodiment. The information processing system S is configured by connecting one or more servers 1 arranged in one or more user environments and multiple nodes 21 installed in multiple bases 2 located in different regions to each other via a network N.

[0018] The server 1 transmits a user request input from a user terminal (not shown) to the node 21 via the network N. In addition, the server 1 presents a response from the node 21 to the user request received via the network N to the user via the user terminal.

[0019] The nodes 21 in each base 2 form a multi-node system connected by a back-end LAN (Local Area Network). The storages 22 (four in this embodiment) in each base 2 are object storages as an example of storage areas, and are connected to each node 21 via a SAN (Storage Area Network). The multiple nodes 21 and multiple storages 22 in the same base 2 form a computer system 1S. In this embodiment, storages 22-j correspond to nodes 21-j (j=1, 2, ..., 4), respectively.

[0020] The server 1, the node 21, and the storage 22 may be either on-premise or cloud. The numbers of the server 1, the base 2, the node 21, and the storage 22 are not limited to those shown in the figure. The correspondence between the node 21 and the storage 22 is not limited to one-to-one.

[0021] Each node 21 has, as functional units related to this embodiment, a request management unit 211, a service management unit 212, a metadata management unit 213, a storage management unit (storage area management unit) 214, and a DB (Data Base) 215. Descriptions of other functions of the node 21 will be omitted.

[0022] The request management unit 211 receives a user request such as an I / O request from the server 1, determines a node 21 within the same base 2 that will process the received user request, and transfers the user request to the determined node 21.

[0023] The service management unit 212 performs anonymization of the custom metadata assigned to the actual data by the user. The actual data is the main data stored in the storage 22 by the user, and the type and format of the data are not important.

[0024] The metadata management unit 213 stores and manages metadata that has been systematically assigned to actual data in the BD 215. The metadata management unit 213 also stores and manages metadata that has been systematically assigned to anonymized custom metadata, which is obtained by anonymizing custom metadata, in the BD 215. Metadata management information 215T in the BD 215 that manages metadata will be described later with reference to FIG. 6.

[0025] The storage management unit 214 stores the actual data, the custom metadata, and the object including the metadata in the storage 22. The storage management unit 214 also stores the object including the anonymized custom metadata and the metadata in the storage 22.

[0026] 1, an object including actual data 101A, custom metadata 102A of the actual data 101A, and first metadata 103A is referred to as a first object 100A. The first metadata 103A is metadata that is assigned to the actual data 101A.

[0027] An object including dummy data 101B of real data 101A, anonymized custom metadata 102B obtained by anonymizing custom metadata 102A, and second metadata 103B is called a second object 100B. The second metadata 103B is metadata added to the anonymized custom metadata 102B. The dummy data 101B may be null data or a virtual representation of non-existent data.

[0028] When instructed to store the first object 100A and the second object 100B, the storage management unit 214 stores the first object 100A and the second object 100B in the storage 22 managed by the storage management unit 214 itself.

[0029] (Object storage region) 2 is an explanatory diagram of an object storage region. FIG. 3 is a diagram showing a region map 2151.

[0030] In this embodiment, the first metadata 103A and the second metadata 103B are stored and managed in regions obtained by dividing the entire storage area of ​​the DB 215 of all nodes 21 in the same base 2 into a predetermined number (32 in this embodiment). Each region includes a primary region and a secondary region that is located in a node 21 different from the primary region. The region map 2151 is information that indicates the node 21 in which the primary region and secondary region are located for each region identified by a region number.

[0031] The first metadata 103A is stored in a pair of the primary region and secondary region. The region (region number) in which the first metadata 103A is stored is determined by a hash value calculated based on path information of the storage destination designated by the user. The first metadata 103A identifies the node 21 in which the primary region and secondary region are located based on the determined region number and the region map 2151. Then, the first metadata 103A is stored in the primary region and secondary region of the DB 215 of each identified node 21.

[0032] The management of the first metadata 103A and the second metadata 103B is not limited to the management method using regions, and various management methods can be adopted.

[0033] (Data storage process) Fig. 4 is a sequence diagram showing an object storage process according to the embodiment. Fig. 4 shows an example in which the first metadata 103A is stored in the DB 215 of the node 21-2 (node ​​#2) as the primary region and the node 21-3 (node ​​#3) as the secondary region in the base 2-1. Also, an example is shown in which the first object 100A is stored in the storage 22-3 corresponding to the node 21-3 (node ​​#3).

[0034] First, in step S11, the request management unit 211 of the node 21-1 receives the actual data 101A and the custom metadata 102A (FIG. 1) of the actual data 101A together with a storage request from the server 1. The node 21 to which various requests are to be sent is determined using a load balancing method such as DNS (Domain Name System) round robin.

[0035] Next, in step S12, the request management unit 211 of the node 21-1 creates first metadata 103A for the first object 100A including the actual data 101A and the custom metadata 102A received in step S11. The first metadata 103A includes the creator (user name) of the actual data 101A, the creation date, the update date, storage destination information, the size, the hash value of the actual data 101A, and other information.

[0036] Here, the metadata 103 (first metadata 103A, second metadata 103B) will be described. FIG. 5 is a diagram showing the configuration of the metadata 103 according to this embodiment. The metadata 103 includes, as an example, items such as EF (External Files), IF (Internal Files), ECM (External Custom Metadata), ICM (Internal Custom Metadata), Masking, MCM (Masked Custom Metadata), and Replication. Of these, EF, IF, and Replication are the same items as in the past, and ECM, ICM, Masking, and MCM are items added in this embodiment.

[0037] ECM is a path that can be specified by the user, and is obtained by adding the file name of the custom metadata 102A to the path value of EF. ICM is a path that internally manages the custom metadata 102A. When an object including the custom metadata 102A is stored in the storage 22 for the first time, the ICM is obtained by adding the file name of the custom metadata 102A to the path value of IF. Also, when an object including the custom metadata 102A or the anonymized custom metadata 102B is stored in the storage 22 for the second or subsequent time, the ICM is obtained by adding the file name of the custom metadata 102A to the object ID of the object.

[0038] Masking is a flag indicating whether or not to perform anonymization processing of the custom metadata 102A, and is set by the user. MCM is a path for internally managing anonymized custom metadata 102B obtained by anonymizing the custom metadata 102A. When an object including the custom metadata 102A is stored in the storage 22 for the first time, the MCM is set to "null" since the anonymized custom metadata 102B does not exist. Moreover, when an object including the custom metadata 102A or the anonymized custom metadata 102B is stored in the storage 22 for the second or subsequent time, the MCM is set to the same value as the ICM of the anonymized custom metadata 102B.

[0039] Returning to the explanation of Fig. 4, next in step S13, the request management unit 211 of the node 21-1 calculates a hash value of the data storage path EF (Fig. 5) designated by the user, and determines the region of DB 215 in which the actual data 101A received in step S11 is to be stored. Next, in step S14, the request management unit 211 of the node 21-1 transmits the first metadata 103A to each node 21 having the primary region and secondary region of the storage destination DB 215 determined in step S13. Here, it is assumed that the primary region of region 2 in which the first metadata 103A is stored exists in node 21-2 (node ​​#2), and the secondary region exists in node 21-3 (node ​​#3).

[0040] Next, in step S15, the metadata management unit 213 of node 21-2 adds the first metadata 103A received from node 21-1 to the metadata management information 215T of the specified primary region in the DB 215 of its own node. In step S16, the metadata management unit 213 of node 21-3 adds the first metadata 103A received from node 21-1 to the metadata management information 215T of the specified secondary region in the DB 215 of its own node, similar to step S15.

[0041] Here, the metadata management information 215T will be described. Fig. 6 is a diagram showing the metadata management information 215T. As shown in Fig. 6, a record 215T1 is a record of the first metadata 103A that is added to the metadata management information 215T of the authentic region in step S15.

[0042] Record 215T1 includes EF “ / dirA / dirB / fileA”, IF “node#2: / ObjectA / fileA”, ..., ECM “ / dirA / dirB / metadata.xml”, ICM “node#2: / ObjectA / metadata.xml”, ..., Masking “true”, MCM “node#2: / ObjectB / metadata.xml”, ..., Replication “Site#2: / node#3: / ObjectC / fileA” as field values ​​of “ObjectA”, which is the first object 100A including the first metadata 103A.

[0043] Next, in step S17, the request management unit 211 of the node 21-1 calculates the free space of the storage 22 of each node 21, and determines the storage 22 in which to store the first object 100A (Object A). Here, it is assumed that the storage 22-3 of the node 21-3 (node ​​#3) is determined as the storage destination.

[0044] Next, in step S18, the request management section 211 of the node 21-1 transmits the first object 100A to the node (node ​​21-3) in charge of the storage destination 22-3 determined in step S17.

[0045] Next, in step S19, the storage management section 214 of the node 21-3 stores the received first object 100A in the storage 22-3.

[0046] (Custom Metadata Anonymization Processing According to the Embodiment) Fig. 7 is a sequence diagram showing a custom metadata anonymization process according to the embodiment. Fig. 7 shows an example of the custom metadata anonymization process executed on the first object 100A after it is stored in the storage 22-3 in step S19 of the process when storing the object.

[0047] The custom metadata anonymization process is executed by a plurality of nodes 21 in cooperation with each other after the object storage process (FIG. 4) of the first object 100A.

[0048] First, in step S21, the service management unit 212 of the node 21-1 transmits a record creation request for the second metadata 103B to the nodes 21-2 and 21-3 having a region that manages the first metadata 103A of the first object 100A. Here, the node 21-1 is the node 21 that has been selected as having sufficient resources (CPU and memory) to issue a record creation request for the second metadata 103B of the second object 100B.

[0049] Next, in step S22, the metadata management unit 213 of node 21-2 creates second metadata 103B for the second object 100B based on the first metadata 103A for the first object 100A (Object A) in the metadata management information 215T of its own node. Then, the metadata management unit 213 adds a record of the created second metadata 103B to the metadata management information 215T of its own node.

[0050] For example, as illustrated in FIG. 6, record 215T2 is a record of second metadata 103B added based on record 215T1 of first metadata 103A in metadata management information 215T of region 2, which is the authentic region. When Masking of record 215T1 of first metadata 103A is "true" and MCM is "null", record 215T2 of second metadata 103B that manages anonymous custom metadata 102B is added to the same region 2 based on record 215T1. At this time, the same value as the internal path IF of the actual data of "ObjectA" is set to the internal path IF of the actual data of "ObjectB". Also, the same value as the ICM of "ObjectB" is set to the MCM of "ObjectA".

[0051] In step S23, the metadata management unit 213 of the node 21-3 creates second metadata 103B for the second object 100B based on the first metadata 103A for the first object 100A, "Object A," in the metadata management information 215T of the own node. Then, the metadata management unit 213 adds a record of the created second metadata 103B to the metadata management information 215T of the own node.

[0052] Next, in step S24, the metadata management unit 213 of node 21-2 transmits an ICM indicating the storage location of the custom metadata 102A of the first object 100A (Object A) and the second metadata 103B of the second object 100B (Object B) to the service management unit 212 of the requesting node 21-1.

[0053] Next, in step S25, the service management unit 212 of the node 21-1 requests the node 21-3, which manages the storage 22-3 in which the custom metadata 102A of the first object 100A (Object A) is stored, to acquire the custom metadata 102A.

[0054] Next, in step S26, the storage management unit 214 of the node 21-3 obtains the custom metadata 102A of the first object 100A (Object A) from the storage 22-3, and transmits it to the service management unit 212 of the node 21-1.

[0055] Next, in step S27, the service management unit 212 of the node 21-1 creates anonymized custom metadata 102B by anonymizing the custom metadata 102A of the first object 100A (ObjectA) received from the node 21-3. Anonymization is a process of processing information until details cannot be identified, such as deleting or replacing at least a part of the information to a level where an individual cannot be identified, when the custom metadata 102 includes personal information, for example. Note that in this embodiment, anonymization is used as an example of a method for protecting the information of the custom metadata 102A, but is not limited to anonymization, and other information protection processing methods such as concealment and the like can also be adopted.

[0056] Next, in step S28, the service management unit 212 of node 21-1 determines the storage 22 in which to store the anonymized custom metadata 102B. Here, it is assumed that the storage 22-4 is determined as the storage destination based on the available space. Next, in step S29, the service management unit 212 of node 21-1 transmits the second object 100B including the second metadata 103B and the anonymized custom metadata 102B to the node 21-4 in charge of the storage 22-4 determined in step S28.

[0057] Next, in step S30, the storage management section 214 of the node 21-4 stores the received second object 100B in the storage 22-4.

[0058] (Data access processing) Fig. 8 is a sequence diagram showing data access processing according to the embodiment. Fig. 8 shows an example in which whether to respond with actual data 101A and custom metadata 102A or actual data 101A and anonymized custom metadata 102B is switched depending on the custom metadata reference authority of the user who has made a data reference request from the server 1.

[0059] It is assumed that, prior to the execution of the data access process, the presence or absence of a user's data reference authority is defined in advance for each of the first object 100A, the actual data 101A, and the custom metadata 102A, and is stored in each node 21. It is also assumed that, prior to the execution of the data access process, the custom metadata anonymization process shown in Fig. 7 has been executed, the first object 100A has been stored in the storage 22-3 in the base 2-1, and the second object 100B has been stored in the storage 22-4.

[0060] First, in step S31, the request management unit 211 of the node 21-1 receives a reference request for the first object 100A by a user from the server 1. The node 21-1 is determined as the node 21 that will receive the reference request by a load balancing method such as DNS round robin.

[0061] Next, in step S32, the request management unit 211 of the node 21-1 determines a region in which the first metadata 103A of the reference target first object 100A is managed based on the hash value of the path (EF) designated by the user. Then, the request management unit 211 identifies the node 21 that manages the authentic region based on the determined region and the region map 2151. Here, it is assumed that the node 21-2 is identified.

[0062] Next, in step S33, the request management section 211 of the node 21-1 transmits an information request for the first metadata 103A of the first object 100A registered in the metadata management information 215T to the node 21-2 identified in step S32.

[0063] Next, in step S34, the metadata management section 213 of the node 21-2 transmits the information of the first metadata 103A of the first object 100A registered in the metadata management information 215T to the node 21-1.

[0064] Next, in step S35, the metadata management section 213 of node 21-2 requests node 21-3 storing first object 100A to acquire and transmit first object 100A based on the information of first metadata 103A transmitted in step S34. Here, it is assumed that node 21-3 storing first object 100A has been identified based on the information of first metadata 103A transmitted in step S34.

[0065] Next, in step S36, the storage management section 214 of node 21-3 acquires the first object 100A requested for acquisition and transmission from storage 22-3, and transmits it to node 21-1.

[0066] Next, in step S37, request management unit 211 of node 21-1 determines whether the user who sent the reference request in step S31 has reference authority to first object 100A sent in step S36. If the user who sent the reference request in step S31 has reference authority to first object 100A (step S37 YES), request management unit 211 proceeds to step S38, and if not (step S37 NO), proceeds to step S39.

[0067] In step S38, the request management unit 211 of the node 21-1 responds to the server 1 with the actual data 101A and the anonymized custom metadata 102B included in the first object 100A received from the node 21-3.

[0068] Meanwhile, in step S39, the request management unit 211 of node 21-1 sends an information request to node 21-2 identified in step S32 for the second metadata 103B of the second object 100B that is associated with the first metadata 103A of the first object 100A in the metadata management information 215T.

[0069] Next, in step S40, the metadata management section 213 of the node 21-2 transmits information on the second metadata 103B of the second object 100B, which is registered in the metadata management information 215T, to the node 21-1.

[0070] Next, in step S41, the metadata management unit 213 of node 21-2 requests node 21-4, which stores the second object 100B, to acquire and transmit the second object 100B, based on the information of the second metadata 103B transmitted in step S40. Here, it is assumed that node 21-4, which stores the second object 100B, has been identified based on the information of the second metadata 103B transmitted in step S40.

[0071] Next, in step S42, the storage management section 214 of the node 21-4 acquires the second object 100B requested to be acquired and transmitted from the storage 22-4, and transmits it to the node 21-1.

[0072] Next, in step S43, the request management unit 211 of the node 21-1 responds to the server 1 with the anonymized custom metadata 102B contained in the second object 100B received from the node 21-4, together with the actual data 101A contained in the first object 100A that has already been acquired.

[0073] (Object update process) The operation of the update process for objects (actual data within objects) differs depending on whether the object storage's generation management function (versioning) is enabled or disabled. Below, we will explain the process when versioning is disabled and when versioning is enabled separately.

[0074] (Object update process when versioning is disabled) Fig. 9A is a sequence diagram showing an object update process (when versioning is disabled) according to the embodiment. Fig. 9A shows an example of updating actual data 101A with actual data for update in response to an update request for actual data 101A made by a user from server 1 when versioning is disabled. It is assumed that a first object 100A has been stored in storage 22-3 and a second object 100B has been stored in storage 22-4 in base 2-1 prior to execution of the object update process (when versioning is disabled).

[0075] First, in step S51, the request management unit 211 of the node 21-1 receives an update request for the actual data 101A included in the first object 100A by a user from the server 1. The node 21-1 is the node 21 that has been determined to receive the update request by a load balancing method such as DNS round robin.

[0076] Next, in step S52, the request management unit 211 of the node 21-1 determines a region in which the first metadata 103A of the first object 100A to be updated is managed, based on the hash value of the path (EF) designated by the user. Then, the request management unit 211 identifies the node 21 that manages the authentic region, based on the determined region and the region map 2151. Here, it is assumed that the node 21-2 is identified.

[0077] Next, in step S53, the request management section 211 of the node 21-1 transmits an information request for the first metadata 103A of the first object 100A registered in the metadata management information 215T to the node 21-2 identified in step S52.

[0078] Next, in step S54, the metadata management section 213 of the node 21-2 transmits the information of the first metadata 103A of the first object 100A registered in the metadata management information 215T to the node 21-1.

[0079] Next, in step S55, the metadata management section 213 of node 21-2 transmits an update request for actual data 101A included in first object 100A to node 21-3 storing first object 100A based on the information of first metadata 103A transmitted in step S54. Here, it is assumed that node 21-3 storing first object 100A has been identified based on the information of first metadata 103A transmitted in step S54.

[0080] Next, in step S56, the storage management unit 214 of the node 21-3 updates the actual data 101A included in the first object 100A for which the update request has been made and which is stored in the storage 22-3, with the actual data for update.

[0081] (Object update process when versioning is enabled) Fig. 9B is a sequence diagram showing an object update process (when versioning is enabled) according to the embodiment. Fig. 9B shows an example of updating the actual data 101A in a generationally manageable manner (i.e., storing both the actual data 101A before the update and the actual data for update) in response to an update request for the actual data 101A made by a user from the server 1 when versioning is enabled. It is assumed that, prior to execution of the object update process (when versioning is enabled), the first object 100A has been stored in the storage 22-3 in the base 2-1, and the second object 100B has been stored in the storage 22-4.

[0082] Steps S51B to S53B in FIG. 9B are similar to steps S51 to S53 in FIG. 9A.

[0083] In step S54B, the metadata management section 213 of node 21-2 sets the ECM of the information of the second object 100B registered in the metadata management information 215T in node 21-2 to "null." Next, in step S55B, the metadata management section 213 of node 21-2 sets the EF and ECM of the information of the first object 100A registered in the metadata management information 215T in node 21-2 to "null."

[0084] Next, in step S56B, the nodes 21-1, 21-2, 21-3, and 21-4 execute the object storage process (FIG. 4) for the third object including the actual data for updating the actual data 101A and the custom metadata. By executing step S56B, the third object is stored in any one of the storages 22.

[0085] Next, in step S57B, the nodes 21-1, 21-2, 21-3, and 21-4 execute the custom metadata anonymization process (FIG. 7) for the third object including the actual data for updating the actual data 101A, the custom metadata, and the third metadata. By executing step S57B, a fourth object including the anonymized custom metadata obtained by anonymizing the custom metadata of the third object and the fourth metadata is stored in one of the storages 22.

[0086] In this way, first object 100A and second object 100B before the update are maintained as they are, and every time real data 101A is updated, an object is added and recorded as history, so that generational management of objects can be performed for each update.

[0087] (Object deletion process) Fig. 10 is a sequence diagram showing an object deletion process. Fig. 10 shows an example in which first metadata 103A, second metadata 103B, first object 100A, and second object 100B are deleted in response to an object deletion request from a user received from server 1.

[0088] First, in step S61, the request management unit 211 of the node 21-1 receives a request from the server 1 to delete the first object 100A by a user. The node 21-1 is a node that has been determined to receive the deletion request by a load balancing method such as DNS round robin.

[0089] Next, in step S62, the request management unit 211 of the node 21-1 determines the region in which the first metadata 103A of the first object 100A to be deleted is managed based on the hash value of the path (EF) designated by the user. Then, the request management unit 211 identifies the node 21 that manages the authentic region based on the determined region and the region map 2151. Here, it is assumed that the node 21-2 is identified.

[0090] Next, in step S63, the request management section 211 of the node 21-1 transmits a confirmation request to the node 21-2 identified in step S62 for the information on the first metadata 103A of the first object 100A registered in the metadata management information 215T.

[0091] Next, in step S64, the metadata management unit 213 of node 21-2 checks the presence of second metadata 103B of the second object 100B linked to the first metadata 103A of the first object 100A in the metadata management information 215T. That is, if Masking of the first metadata 103A is "true", the unit 213 starts deleting second metadata 103B having an ICM with the same path value as the MCM of the first metadata 103A in the metadata management information 215T in the same region.

[0092] Next, in step S65, the metadata management section 213 of the node 21-2 transmits information on the second metadata 103B of the second object 100B, the deletion of which has started in step S64, to the request management section 211 of the node 21-1.

[0093] Next, in step S66, the metadata management section 213 of node 21-2 requests node 21-4, which stores second object 100B to which the information of second metadata 103B was transmitted in step S65, to delete second object 100B. Here, it is assumed that node 21-4, which stores second object 100B, has been identified based on the information of second metadata 103B transmitted in step S65.

[0094] Next, in step S67, the storage management unit 214 of the node 21-4 deletes the second object 100B for which the deletion request has been made from the storage 22-4.

[0095] Next, in step S68, the request management section 211 of the node 21-1 notifies the metadata management section 213 of the node 21-2 of the completion of the deletion of the second object 100B.

[0096] Next, in step S69, the metadata management section 213 of node 21-2 deletes the record of second metadata 103B of second object 100B in metadata management information 215T. Then, the metadata management section 213 sets Masking of first metadata 103A of first object 100A to "false" and MCM to "null", and starts deleting first metadata 103A.

[0097] Next, in step S70, the metadata management section 213 of the node 21-2 transmits information on the first metadata 103A of the first object 100A, the deletion of which has started in step S69, to the request management section 211 of the node 21-1.

[0098] Next, in step S71, the request management unit 211 of node 21-1 requests node 21-3, which stores the first object 100A to which the information of the first metadata 103A was transmitted in step S70, to delete the first object 100A. Here, it is assumed that node 21-3 storing the first object 100A has been identified based on the information of the first metadata 103A transmitted in step S70.

[0099] Next, in step S72, the storage management section 214 of the node 21-3 deletes the first object 100A for which the deletion request has been made from the storage 22-3.

[0100] Next, in step S73, the request management section 211 of node 21-1 notifies the metadata management section 213 of node 21-2 of the completion of deletion of the first object 100A. Next, in step S74, the metadata management section 213 of node 21-2 deletes the record of the first metadata 103A of the first object 100A in the metadata management information 215T.

[0101] (Data corruption handling) Fig. 11 is a sequence diagram showing the process when data is corrupted. The process when data is corrupted shows the process when the first object 100A that the user is trying to refer to via the server 1 is corrupted. In Fig. 11, the primary region in which the first metadata 103A of the reference target first object 100A is managed exists in the node 21-2, and the first object 100A is stored in the storage 22-3 of the node 21-3. Also, a first object (replica object) 100C which is a replica of the first object 100A is saved in the replication environment of the primary environment built in the replication destination base 2-2 (see Fig. 1).

[0102] If the first object 100A stored in the storage 22-3 of the base 2-1, which is the primary environment, is corrupted when read, the first object 100C stored in the storage 22-3 of the base 2-2, which is the secondary environment, is acquired. Also, the first metadata 103A is assumed to hold in advance the hash value of the first object 100A.

[0103] First, in step S81, the request management unit 211 of the node 21-1 determines a region in which the first metadata 103A of the first object 100A to be referenced is managed, based on a hash value of a path (EF) specified in a user reference request received from the server 1. Then, the request management unit 211 identifies the node 21-2 that manages the authentic region, based on the determined region and the region map 2151. Here, it is assumed that the node 21-2 has been identified. Then, the request management unit 211 transmits an information request for the first metadata 103A of the first object 100A to the identified node 21-2.

[0104] Next, in step S82, the metadata management section 213 of the node 21-2 reads information of the first metadata 103A of the first object 100A (Object A) that is the request target from the DB 215 of the node 21-2, and transmits it to the request management section 211 of the node 21-1.

[0105] Next, in step S83, the request management unit 211 of the node 21-1 requests the node 21-3 connected to the storage 22-3 in which the first object 100A (Object A) is saved to acquire the first object 100A (Object A). Here, it is assumed that the node 21-3 storing the first object 100A (Object A) is identified based on the information of the first metadata 103A transmitted in step S82.

[0106] Next, in step S84, the storage management section 214 of the node 21-3 obtains the first object 100A (Object A) from the storage 22-3, and transmits it to the request management section 211 of the node 21-1.

[0107] Next, in step S85, the request management unit 211 of the node 21-1 calculates the hash value of the first object 100A transmitted in step S84. The request management unit 211 then compares the hash value of the first object 100A (ObjectA) held in the first metadata 103A transmitted in step S82 with the calculated hash value of the first object 100A. If the hash value of the first object 100A held in the first metadata 103A and the calculated hash value are different (abnormal) (step S85 YES), the request management unit 211 proceeds to step S86. On the other hand, if they match (normal) (step S85 NO), the request management unit 211 responds to the server 1 with the first object 100A, and ends this data corruption processing.

[0108] In step S86, the request management unit 211 of the node 21-1 refers to the first metadata 103A and checks the region of the replication environment in which the first object 100C (Object C), which is a copy of the first object 100A (Object A), is managed. Then, the request management unit 211 connects to the region in which the first object 100C (Object C) of the replication environment (site 2-1) is managed via an inter-site connection.

[0109] Next, in step S87, the request management unit 211 of the node 21-1 at the base 2-2 determines a region in which the first metadata 103C of the first object 100C to be connected is managed based on a hash value of the path (EF) designated by the user. The node 21-1 is a node 21 determined using a load balancing method such as DNS round robin. Then, the request management unit 211 of the node 21-1 specifies the node 21-2 that manages the authentic region based on the determined region and the region map 2151. Here, it is assumed that the node 21-2 is specified. Then, the request management unit 211 of the node 21-1 at the base 2-2 transmits an information request for the first metadata 103C of the first object 100C to the specified node 21-2.

[0110] Next, in step S88, in the base 2-2, the metadata management unit 213 of the node 21-2 reads information of the first metadata 103C of the first object 100C (Object C) that is the subject of the information request from the DB 215 of the node 21-2. The metadata management unit 213 transmits the information of the first metadata 103C to the request management unit 211 of the node 21-1.

[0111] Next, in step S89, the request management unit 211 of the node 21-1 requests the node 21-3 connected to the storage 22-3 in which the first object 100C (Object C) is stored to acquire the first object 100C (Object C). Here, it is assumed that the node 21-3 storing the first object 100C (Object C) is identified based on the information of the first metadata 103C transmitted in step S88.

[0112] Next, in step S90, the storage management section 214 of the node 21-3 obtains the first object 100C (Object C) from the storage 22-3, and transmits it to the request management section 211 of the node 21-1.

[0113] Next, in step S91, the request management unit 211 of node 21-1 transfers the first object 100C (Object C) received from node 21-3 to which storage 22-3 is connected to the request management unit 211 of node 21-1 at site 2-1 which is the request source. The request management unit 211 of node 21-1 at site 2-1 responds to the server 1 with the first object 100C (Object C) received from the request management unit 211 of node 21-1 at site 2-2 as a substitute for the first object 100A (Object A).

[0114] Even if the custom metadata 102A is anonymized in this way, the replication function can be used in the same way as in conventional object storage.

[0115] (Effects of the embodiment) In the above embodiment, the actual data 101A is managed only by the first object 100A, and the second object 100B including the anonymized custom metadata 102B refers to the actual data of the first object 100A. Therefore, when the first object 100A before the anonymization of the custom metadata is updated, the update can be reflected in the second object 100B after the anonymization of the custom metadata at the same time, without complicating data management. In addition, since the second object 100B including the anonymized custom metadata 102B is managed independently of the first object 100A, only the anonymized custom metadata 102B can be deleted by deleting the second object 100B.

[0116] Furthermore, by managing the first metadata 103A and the second metadata 103B in the metadata management information 215T of the same region, it is possible to search for mutual information in the metadata of the first object 100A and the second object 100B. Furthermore, by providing a "Masking" flag to the metadata 103 and referring to the metadata management information 215T that manages the metadata 103 when updating the first object 100A, it is possible to perform an update process for the second object 100B linked to the first object 100A.

[0117] Furthermore, when the second object 100B is updated after the object update process is executed, a record for managing new second metadata 103B for the second object 100B is added to the metadata management information 215T. In this way, the history of the second metadata 103B before the update is recorded, so that generation management of the first object 100A and the second object 100B can be performed. In this way, in the embodiment, it is possible to use functions of conventional object storage (such as versioning and replication).

[0118] (Modification of the embodiment) 12 is an explanatory diagram of a modified example of the embodiment. In the above-described embodiment, the anonymized custom metadata 102B is included in the second object 100B including the dummy data 101B and the second metadata 103B, and is managed in an object different from the first object 100A including the first metadata 103A.

[0119] However, the method of managing the anonymous custom metadata 102B is not limited to that described in the embodiment. That is, as a first modification, the anonymous custom metadata 102B may be included in the second object 100B together with the actual data 101A and the second metadata 103B, and managed in an object different from the first object 100A.

[0120] Modification 1 differs from the custom metadata anonymization process (FIG. 7) described in the embodiment in the following respects. That is, when creating second metadata 103B of second object 100B, the storage path of first object 100A is not included in second metadata 103B as a reference path of actual data 101A. Also, in the custom metadata anonymization process, when second object 100B is stored in storage 22, actual data 101A is included in second object 100B.

[0121] Moreover, in the first modification, the following points are different from the data access process (FIG. 8) described in the embodiment. That is, in response to a reference request from the server 1, if the user has the reference authority to the custom metadata 102A, the first object 100A is acquired, and the actual data 101A and the custom metadata 102A of the first object 100A are returned to the server 1. On the other hand, in response to a reference request from the server 1, if the user does not have the reference authority to the custom metadata 102A, the second object 100B is acquired, and the actual data 101A and the anonymized custom metadata 102B of the second object 100B are returned to the server 1.

[0122] Moreover, in the first modification, the following points are different from the object update process (FIGS. 9A and 9B) described in the embodiment: When real data 101A is updated, the update is reflected in both first object 100A and second object 100B.

[0123] Furthermore, in variant example 1, unlike the processing when data is corrupted described in the embodiment (FIG. 11), when the first object 100A is corrupted, the actual data 101A may be obtained from the second object 100B stored within the base 2 where the first object 100A is stored.

[0124] Other than the above differences, the processing in the first modification is the same as that in the embodiment.

[0125] Alternatively, as a second modification, the anonymized custom metadata 102B may be added to and managed in the same first object 100A as the real data 101A, the custom metadata 102A, and the first metadata 103A.

[0126] Modification 2 differs from the custom metadata anonymization process (FIG. 7) described in the embodiment in the following respects: That is, a second object 100B is not created, and anonymized custom metadata 102B is added to a first object 100A.

[0127] Moreover, in the second modification, the following points are different from the data access process (FIG. 8) described in the embodiment: In response to a reference request from the server 1, if the user does not have the permission to reference the custom metadata 102A, the actual data 101A of the first object 100A and the anonymized custom metadata 102B are returned to the server 1.

[0128] Furthermore, in variant example 2, compared to the object deletion process (FIG. 10) described in the embodiment, since second object 100B does not exist, the deletion process of second object 100B is not performed, and only the deletion process of first object 100A is performed.

[0129] Other than the above differences, the processing in the second modification is the same as that in the embodiment.

[0130] (Computer 1000 Hardware) FIG. 13 is a hardware diagram of a computer 1000 that realizes each node 21. As shown in FIG.

[0131] The computer 1000 includes a processor 1001 including a CPU, a main memory device 1002, an auxiliary memory device 1003, a network interface 1004, an input device 1005, and an output device 1006, all of which are interconnected via an internal communication line 1009 such as a bus.

[0132] The processor 1001 controls the overall operation of the computer 1000. The main memory device 1002 is composed of, for example, a volatile semiconductor memory, and is used as a work memory for the processor 1001. The auxiliary memory device 1003 is composed of a large-capacity nonvolatile memory device such as a hard disk device, an SSD (Solid State Drive), or a flash memory, and is used to store various programs and data for a long period of time.

[0133] An executable program 1100 stored in the auxiliary storage device 1003 is loaded into the main storage device 1002 when the computer 1000 is started up or when necessary, and is executed by the processor 1001 to realize various systems that perform various processes.

[0134] The executable program 1100 may be recorded on a non-transitory recording medium, read from the non-transitory recording medium by a media reading device, and loaded into the primary storage device 1002. Alternatively, the executable program 1100 may be obtained from an external computer via a network, and loaded into the primary storage device 1002.

[0135] The network interface 1004 is an interface device for connecting the computer 1000 to each network in the system or for communicating with other computers. The network interface 1004 is, for example, configured with a NIC (Network Interface Card) for a wired LAN (Local Area Network) or a wireless LAN.

[0136] The input device 1005 is composed of a keyboard, a pointing device such as a mouse, and the like, and is used by the user to input various instructions and information to the computer 1000. The output device 1006 is composed of a display device such as a liquid crystal display or an organic EL (Electro Luminescence) display, and an audio output device such as a speaker, and is used to present necessary information to the user when necessary.

[0137] Although one embodiment of the present invention has been described above in detail, the present invention is not limited to the above embodiment, and various modifications can be made without departing from the spirit of the present invention. For example, the above embodiment has been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to the embodiment having all of the configurations described above. In addition, it is possible to add, delete, or replace part of the configuration of the above embodiment with other configurations.

[0138] In addition, each of the configurations, functional units, processing units, processing means, etc. in the above-mentioned embodiments may be realized in part or in whole by hardware, for example, by designing them as integrated circuits. Each of the above-mentioned configurations, functions, etc. may be realized in software by a processor interpreting and executing a program that realizes each function. Information such as the program, table, file, etc. that realizes each function can be stored in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), an IC card, an SD card, a DVD, or other recording medium.

[0139] In addition, in each of the drawings of the above-mentioned embodiments, the control lines and information lines are shown as those considered necessary for the explanation, and not all of the control lines and information lines in the actual implementation are shown. For example, it may be considered that almost all of the components are actually connected to each other.

[0140] Furthermore, the arrangement of each function and data of the information processing system S described in the above embodiment is merely one example. The multiple nodes 21 and storage 22 in each base 2 of the information processing system S constitute a multi-node system, but configurations other than a multi-node system are also possible. Furthermore, the arrangement of the functions and data of the information processing system S can be changed to an optimal arrangement in terms of hardware and software performance, processing efficiency, communication efficiency, etc. Furthermore, the order of execution of each processing step described in the above embodiment can be changed as long as the processing result is the same.

[0141] Furthermore, the configuration (schema, etc.) of the database that stores the various types of data described above can be flexibly changed from the perspective of efficient use of resources, improved processing efficiency, improved access efficiency, improved search efficiency, and the like. [Explanation of symbols]

[0142] S: information processing system, 1: server, 2, 2-1, 2-2: bases, 21, 21-1, 21-2, 21-3, 21-4: nodes, 22, 22-1, 22-2, 22-3, 22-4: storage, 100A: first object, 101A: actual data, 102A: custom metadata, 103A: first metadata, 100B: second object, 101B: dummy data, 102B: anonymized custom metadata, 103B: second metadata, 211: request management unit, 212: service management unit, 213: metadata management unit, 214: storage management unit, 215: DB, 215T: metadata management information, 1000: computer, 1001: processor, 1002: main memory device.

Claims

1. 1. An object management method executed by a computer system having a storage area in which objects are stored, comprising: the computer system includes a processor and a database in cooperation with a memory; The processor, storing, in the storage area, a first object including actual data to which custom metadata, which is metadata customizable by a user, is added, in the storage area in response to a storage request for storing the actual data in the storage area, the custom metadata being customizable by a user; creating anonymized custom metadata by anonymizing the custom metadata; storing a second object in the storage area, the second object including the anonymized custom metadata; In response to the storage request, a first metadata including storage destination information of the first object is generated and registered in the database; storing the first object in the storage area including the first metadata; acquiring the custom metadata from the storage area based on the first metadata registered in the database; The second object, which does not include the actual data but includes the anonymized custom metadata, is stored in the storage area. The object management method includes each of the processes.

2. 2. The object management method according to claim 1, The processor, creating second metadata having storage destination information of the first object including the actual data and storage destination information of the second object, and registering the second metadata in the database; The second object is stored in the storage area while including the second metadata in the second object. The object management method includes each of the processes.

3. 3. The object management method according to claim 2, further comprising: The processor, In response to a reference request for referencing the first object, verify the authority of a requesting user who issued the reference request; If the requesting user has an authority to reference the custom metadata, retrieve the first object from the storage area based on the first metadata registered in the database, and respond to the reference request with the retrieved first object; When the requesting user does not have the authority to refer to the custom metadata, the actual data and the anonymized custom metadata are acquired from the storage area based on the second metadata registered in the database, and the access request is responded to with an object including the acquired actual data and the anonymized custom metadata. The object management method includes each of the processes.

4. 3. The object management method according to claim 2, further comprising: The first metadata includes storage location information of the second object including the anonymized custom metadata.

13. An object management method comprising:

5. 5. An object management method according to claim 4, further comprising the steps of: The processor, In response to a deletion request for deleting the first object, the second metadata having the same value as storage destination information of the anonymized custom metadata included in the first metadata in the database is identified, and the second object stored in the storage area is deleted based on the identified second metadata; The first object stored in the storage area is deleted based on the first metadata registered in the database. The object management method includes each of the processes.

6. 6. An object management method according to claim 5, further comprising the steps of: The processor, The second object is deleted first, and the first object is deleted after the deletion of the second object is completed.

13. An object management method comprising:

7. 2. The object management method according to claim 1, The processor, registering, in the database, a hash value of the actual data and storage destination information in a replication destination computer system that stores a replicated object obtained by replicating the first object, in the first metadata; In response to a reference request for referencing the first object, the first object is retrieved from the storage area based on the first metadata registered in the database; Calculating a hash value of the acquired first object; comparing the calculated hash value with the hash value registered in the database; if the two compared hash values ​​are different, acquire the replicated object of the first object from a storage area of ​​the replication destination computer system based on the first metadata registered in the database; Responding to the reference request with the replicated object obtained from the replication destination computer system, instead of the first object. The object management method includes each of the processes.

8. 2. The object management method according to claim 1, The processor, upon receiving an update request to update the actual data included in the first object with updated actual data, When a generation management function for the actual data that stores both the actual data and the updated actual data is valid, creating third metadata having storage destination information of a third object including the updated actual data and the custom metadata, and registering the third metadata in the database; storing the third object in the storage area including the third metadata; acquiring the custom metadata from the storage area based on the third metadata registered in the database; creating anonymized custom metadata by anonymizing the custom metadata acquired from the storage area based on the third metadata; storing, in the storage area, a fourth object that does not include the updated actual data and that includes the anonymized custom metadata obtained by anonymizing the custom metadata acquired from the storage area based on the third metadata; If the generation management function is disabled, updating the first object with the updated actual data based on the first metadata registered in the database; The object management method includes each of the processes.

9. A computer system having a storage area and a database in which objects are stored, a storage area management unit that, in response to a storage request for storing actual data to which custom metadata, which is metadata customizable by a user, is added in the storage area, stores a first object including the actual data and the custom metadata in the storage area; a service management unit that creates anonymized custom metadata by anonymizing the custom metadata, The storage area management unit storing a second object in the storage area, the second object including the anonymized custom metadata; In response to the storage request, a first metadata including storage destination information of the first object is generated and registered in the database; storing the first object in the storage area including the first metadata; acquiring the custom metadata from the storage area based on the first metadata registered in the database; The second object, which does not include the actual data but includes the anonymized custom metadata, is stored in the storage area.

1. A computer system comprising:

10. An object management program for causing a computer to function as a computer system having a memory area and a database in which objects are stored, comprising: The computer includes: storing, in the storage area, a first object including actual data to which custom metadata, which is metadata customizable by a user, is added, in the storage area in response to a storage request for storing the actual data in the storage area, the custom metadata; creating anonymized custom metadata by anonymizing the custom metadata; storing a second object in the storage area, the second object including the anonymized custom metadata; creating first metadata having storage destination information of the first object in response to the storage request, and registering the first metadata in the database; storing the first object in the storage area including the first metadata; acquiring the custom metadata from the storage area based on the first metadata registered in the database; storing the second object in the storage area, the second object not including the actual data, but including the anonymized custom metadata; 2. An object management program comprising:

Citation Information

Patent Citations

  • Document management system and document management method

    JP2019082887A

  • Medical information anonymizing system and anonymization method setting device

    JP2020102214A

  • Anonymization device, and anonymization system

    JP2021051471A

  • Dynamic configuration of patient tags and masking types while de-identifying patient data during image export from PACS diagnostic workstation

    US20040193901A1

  • Multi-Tier Storage Based on Data Anonymization

    US20160132697A1