WO2021135280A1 - Procédé de contrôle de données pour système de stockage distribué, et appareil associé - Google Patents

Procédé de contrôle de données pour système de stockage distribué, et appareil associé Download PDF

Info

Publication number
WO2021135280A1
WO2021135280A1 PCT/CN2020/110952 CN2020110952W WO2021135280A1 WO 2021135280 A1 WO2021135280 A1 WO 2021135280A1 CN 2020110952 W CN2020110952 W CN 2020110952W WO 2021135280 A1 WO2021135280 A1 WO 2021135280A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
target data
crc check
check value
preset
Prior art date
Application number
PCT/CN2020/110952
Other languages
English (en)
Chinese (zh)
Inventor
冯龙
何营
白战豪
Original Assignee
浪潮电子信息产业股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 浪潮电子信息产业股份有限公司 filed Critical 浪潮电子信息产业股份有限公司
Publication of WO2021135280A1 publication Critical patent/WO2021135280A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/08—Error detection or correction by redundancy in data representation, e.g. by using checking codes
    • G06F11/10—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
    • G06F11/1004—Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's to protect a block of data words, e.g. CRC or checksum
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor

Definitions

  • This application relates to the field of distributed storage technology, in particular to a data verification method of a distributed storage system; it also relates to a data verification device, equipment and computer-readable storage medium of a distributed storage system.
  • Mass distributed storage systems usually consist of thousands of hard disks and dozens or hundreds of storage nodes. Each hard disk is managed by an independent storage engine, that is, managed by an independent process. If there are errors in the stored data due to hard disk hardware errors, firmware defects, power supply temperature, etc., it is usually difficult to obtain such anomalies in a timely and effective manner. In a multi-copy distributed storage system, if the wrong data is not found and recovered in time, it may lead to the risk of data loss or data inconsistency. Therefore, it is particularly important to verify the data and find the wrong data in time. At present, the technical solution for data verification is to calculate the CRC (Cyclic Redundancy Check) check value of the data of multiple copies, and compare the CRC check values between the copies.
  • CRC Cyclic Redundancy Check
  • the purpose of this application is to provide a data verification method and related device for a distributed storage system, which can accurately identify the location of data abnormalities and ensure the effective execution of data verification.
  • this application provides a data verification method for a distributed storage system, including:
  • calculating the CRC check value of the data of each preset check interval length in the target data includes:
  • the CRC of the data of each preset check interval length in the target data is directly calculated. Value.
  • calculating the CRC check value of the data of each preset check interval length in the target data includes:
  • start position and/or end position of the write interval of the target data is not an integer multiple of the calculated length of the preset check value, then read the start position from the hard disk before the preset It is assumed that the written data from the position that is an integer multiple of the length of the check interval to the start position, and/or the preset check is read from the end position to the end position from the hard disk The written data at the position that is an integer multiple of the interval length;
  • storing the second CRC check value in the database includes:
  • the second CRC check value and the offset in the object of the data corresponding to the second CRC check value are stored in the database in the form of a map.
  • the reading target data from the hard disk includes:
  • the target data is read from the hard disk when a read request is received or when an active verification is triggered.
  • reading the target data from the hard disk and calculating the CRC check value of the data of each preset check interval length in the target data includes:
  • start position and the end position of the reading interval corresponding to the target data are integer multiples of the length of the preset check interval, directly read and calculate the data of each preset check interval length in the target data CRC check value;
  • start position and/or end position of the read interval of the target data is not an integer multiple of the calculated length of the preset check value, then read the start position from the hard disk before the preset Set the data from the position of an integer multiple of the length of the check interval to the start position and the target data, and/or read the end position to the end position as an integer of the preset check interval length Calculate the CRC check value of the read data and the data of each preset check interval length in the target data.
  • it also includes:
  • the target data is added to the reconstruction queue for data reconstruction.
  • this application also provides a data verification device for a distributed storage system, including:
  • the calculation module is used to read the target data from the hard disk, and calculate the CRC check value of the data of each preset check interval length in the target data to obtain the first CRC check value;
  • a reading module which is used to calculate the second CRC check value obtained by calculating the CRC check value of the data of each preset check interval length in the target data when reading and writing the target data from the database;
  • the comparison module is configured to compare each of the first CRC check values with the corresponding second CRC check values
  • the determining module is configured to determine that an error occurs in the target data in the hard disk if the first CRC check value is inconsistent with the corresponding second CRC check value.
  • this application also provides a data verification device for a distributed storage system, including:
  • Memory used to store computer programs
  • the processor is used to implement the steps of the data verification method of the distributed storage system as described above when the computer program is executed.
  • the present application also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, realizes the above-mentioned distributed storage system.
  • the steps of the data verification method are described in detail below.
  • the data verification method of the distributed storage system includes reading target data from a hard disk, and calculating the CRC check value of the data of each preset check interval length in the target data, to obtain the first CRC check
  • the second CRC check value obtained by calculating the CRC check value of the data of each preset check interval length in the target data; respectively
  • the first CRC check value is compared with the corresponding second CRC check value; if the first CRC check value is inconsistent with the corresponding second CRC check value, it is determined that the hard disk An error occurred in the target data.
  • the data verification method calculates the CRC check value of the data to be written whenever data is written. , And stored in the database.
  • the CRC check value of the data to be checked is calculated, and further compared with the corresponding CRC check value stored in the database, so as to determine whether the data has an error.
  • the data verification method adopts the data verification method in which the storage engine independently performs data verification. Data verification can be completed without using the data in the copies of other nodes. It can not only accurately identify the location of abnormal data, but also It can ensure the effective execution of data verification.
  • the data verification device, equipment, and computer-readable storage medium of the distributed storage system provided in this application all have the above technical effects.
  • FIG. 1 is a schematic flowchart of a data verification method of a distributed storage system provided by an embodiment of the application
  • FIG. 2 is a schematic diagram of calculating a CRC check value provided by an embodiment of the application.
  • FIG. 3 is another schematic diagram of calculating a CRC check value provided by an embodiment of the application.
  • FIG. 4 is a schematic diagram of a data verification device of a distributed storage system provided by an embodiment of the application.
  • FIG. 5 is another schematic diagram of calculating a CRC check value provided by an embodiment of the application.
  • the core of this application is to provide a data verification method and related device for a distributed storage system, which can accurately identify the location of data abnormalities and ensure the effective execution of data verification.
  • FIG. 1 is a schematic flowchart of a data verification method of a distributed storage system according to an embodiment of the application; referring to FIG. 1, the data verification method includes:
  • S101 Read the target data from the hard disk, and calculate the CRC check value of the data of each preset check interval length in the target data to obtain the first CRC check value;
  • this application provides a data verification method that uses a storage engine to independently perform data verification. After starting the data check, read the data that needs to be checked from the hard disk, that is, the above target data, and then calculate the CRC check value of the data of each preset check interval length in the target data, and obtain one or more The first CRC check value. Among them, there is no data duplication among the data in the target data for calculating the CRC check value.
  • the so-called preset check interval length refers to the length of the data for calculating the CRC check value. For example, the preset check value calculation length is 4KB, which means the calculated CRC check value of data with a length of 4KB.
  • the length of the target data is 8KB . Then two first CRC check values can be calculated at this time.
  • the specific value of the length of the preset check interval is not limited in this application, and can be set differently according to actual needs.
  • the specific calculation method for calculating the CRC check value of the data is not described in detail here in this application, and the existing related technology can be referred to.
  • the CRC check value of each preset check interval length in the data to be written is calculated before the data is written to the hard disk to obtain the second CRC check value, and the calculated second CRC check value is obtained.
  • the CRC check value is stored in the database.
  • the second CRC check value calculated and stored in the database when the target data is written is further read from the database, and subsequently based on each first CRC check value and The corresponding second CRC check value determines whether the target data has an error.
  • calculating the CRC check value of the data of each preset check interval length in the target data includes: if the start position and the end of the write interval corresponding to the target data If the position is an integer multiple of the length of the preset check interval, the CRC check value of the data of each preset check interval length in the target data is directly calculated.
  • this embodiment corresponds to the case where the target data is aligned data.
  • the so-called aligned data refers to data in which the start position and the end position of the writing interval in the data buffer are both integral multiples of the length of the preset check interval.
  • the target data is aligned data, at this time, the target data is directly divided by the preset check value calculation length as a unit to obtain multiple data fragments, and the CRC check value of each data fragment is calculated.
  • the start position of the data write interval is 4KB
  • the end position is 16KB
  • the write length is 12KB.
  • the target data is divided into 3 data fragments, and the CRC check value of each data fragment is calculated respectively to obtain 3 first CRC check values: CRC1 to CRC3.
  • calculating the CRC check value of the data of each preset check interval length in the target data includes: if the writing interval of the target data is If the starting position and/or ending position is not an integer multiple of the calculated length of the preset check value, then the position that is an integer multiple of the length of the preset check interval before the starting position is read from the hard disk The written data to the starting position, and/or the written data from the ending position to the position that is an integer multiple of the length of the preset check interval after the ending position is read from the hard disk Input data; Calculate the CRC check value of the data of each preset check interval length in the written data and the target data.
  • this embodiment corresponds to the case where the target data is non-aligned data.
  • the so-called non-aligned data refers to data whose starting position and/or ending position of the writing interval in the data buffer is not an integer multiple of the length of the preset check interval.
  • the target data is unaligned data
  • the data is first read from the hard disk to the data buffer to fill in the unaligned part. Specifically, if the start position of only the write interval is not an integer multiple of the length of the preset check interval, then the position that is an integer multiple of the preset check interval length before the start position is read from the hard disk to the start position The data has been written.
  • the end position of the write-only interval is not an integer multiple of the calculated length of the preset check value
  • the data written from the end position to the position that is an integer multiple of the preset check interval length after the end position is read from the hard disk.
  • the position that is an integer multiple of the preset check interval length before the start position is read from the hard disk to the start position
  • the written data and the written data from the hard disk from the end position to the position that is an integer multiple of the length of the preset check interval after the end position is read from the hard disk.
  • the CRC check value of the data of each preset check interval length is calculated. Among them, it is preferable to read from the hard disk the minimum number of data that can fill in the misaligned part.
  • the start position of the data write interval is 3KB
  • the end position is 15KB
  • the write length is 12KB.
  • the start position and the end position are not integral multiples of the length of the preset check interval, so first read the written data from 0KB to 3KB and the written data from 15KB to 16KB from the hard disk, and then use 0KB
  • the data from position to 16KB is a whole data, and the whole data is divided into 4 data fragments, and the second CRC check value of the data from 0KB to 4KB is calculated, and the second of the data from 4KB to 8KB is calculated.
  • the calculated second CRC check value is further stored in the database, and the target data is written to the hard disk.
  • the storing of the second CRC check value in the database includes: the second CRC check value and the offset of the data corresponding to the second CRC check value in the object are stored in the database in the form of a map.
  • the data on the hard disk includes business data and database data that records index information and metadata information.
  • Business data is stored on the hard disk in units of objects, and the metadata, index information, and extended attributes of the objects are stored in the database.
  • this embodiment adds content called verification data to the object metadata to record the verification information of the data stored on the disk for a certain object.
  • the second CRC check value of the data of each preset check interval length is calculated and recorded in the Key-Value database, where Key is the object name and Value is the object element Data information.
  • the CRC check value of the object and the offset of the data corresponding to the CRC check value in the object are recorded in the database in the form of a map.
  • the organization of the map is as follows:
  • offset characterizes the offset of the data corresponding to the CRC check value in the object.
  • CRC_LENGTH is the length of the preset check interval.
  • the first CRC check value of each data segment of the target data is respectively compared with the first CRC check value of each data segment
  • the two CRC check values are compared; if the first CRC check value of one or some data fragments is inconsistent with the corresponding second CRC check value, it is determined that the target data has an error. On the contrary, if the first CEC check value of each data segment is consistent with the corresponding second CRC check value, there is no error in the target data.
  • the above-mentioned reading of target data from the hard disk includes: reading the target data from the hard disk when a read request is received or when an active verification is triggered.
  • the timing for data verification in this embodiment includes performing data verification on the target data when reading target data (at this time the target data is target read data) after receiving a read request, and triggering active verification. Then, read the target data and perform data verification.
  • the CRC check value of the data of each preset check interval length in the target data including: determining the start of the read interval corresponding to the target data according to the read request Position and end position; if the start position and end position of the reading interval corresponding to the target data are integer multiples of the preset check interval length, directly read and calculate the data of each preset check interval length in the target data CRC check value; if the start position and/or end position of the read interval of the target data is not an integer multiple of the calculated length of the preset check value, the length of the preset check interval is set before the start position is read from the hard disk Data and target data from positions that are integer multiples of the position to the start position, and/or data and target data from positions that are integer multiples of the length of the preset check interval after reading the end position to the end position; calculate the read data and The CRC check value of the data of each preset check interval length in the target data.
  • the target data is aligned data according to the read request. If the target data is aligned data, the data of the corresponding offset length is directly read from the hard disk, and each preset calibration in the target data is calculated. The CRC check value of the data of the length of the check interval. If the target data is non-aligned data, at this time, in addition to reading the target data from the hard disk, other data is also read from the hard disk, so that the target data and the additionally read data together constitute aligned data. Furthermore, the target data and the additionally read data are taken as a whole data, and the CRC check value of each preset check interval length is calculated.
  • the start position of the reading interval corresponding to the target data is 3KB
  • the end position is 15KB
  • the start position and the end position of the read interval are not integer multiples of the length of the preset check interval
  • the 0KB position is also read from the hard disk.
  • the data to the 3KB position, and the data from the 15KB position to the 16KB position, and the data from the 0KB position to the 16KB position as a whole data calculate the CRC check value of the data of each 4KB length.
  • the target data and the additionally read data are taken as a whole data, and the first CRC check value is calculated, and each calculated first CRC check value is related to the corresponding
  • the additional read data is cut from the data buffer, and only the target data is returned to the client.
  • the verification period can be pre-configured, that is, the above-mentioned preset period, for example, data verification is triggered every other week. Therefore, according to the preset period, when the check time is reached, the data check is triggered, the objects in the database are traversed, the target data, which is the target data, is read, and the first CRC check value of the target data is calculated. Read the second CRC check value stored in the database, and compare the second CRC check value with the corresponding first CRC check value.
  • the method of calculating the first CRC check value of the object data reference may be made to the above-mentioned embodiment, which is not repeated in this application.
  • the target data is added to the reconstruction queue for data reconstruction.
  • the data reconstruction process of the distributed storage system is triggered, and the erroneous data is added to the data reconstruction queue as missing data to perform data reconstruction.
  • normal data is read from other copies and written to the local storage engine to restore the local data.
  • the data reconstruction process can use the original reconstruction process of the distributed storage system, which will not be repeated in this application.
  • the data verification method calculates the CRC check value of the data to be written every time data is written, and stores it in the database, and then calculates the data to be verified during subsequent data verification.
  • the CRC check value is further compared with the corresponding CRC check value stored in the database to determine whether the data has an error.
  • the data verification method adopts the data verification method in which the storage engine independently performs data verification. Data verification can be completed without using the data in the copies of other nodes. It can not only accurately identify the location of abnormal data, but also It can ensure the effective execution of data verification.
  • the present application also provides a data verification device of a distributed storage system, and the device described below may correspond to the method described above with reference to each other.
  • the data verification device includes:
  • the calculation module is used to read the target data from the hard disk, and calculate the CRC check value of the data of each preset check interval length in the target data to obtain the first CRC check value;
  • a reading module which is used to calculate the second CRC check value obtained by calculating the CRC check value of the data of each preset check interval length in the target data when reading and writing the target data from the database;
  • the comparison module is configured to compare each of the first CRC check values with the corresponding second CRC check values
  • the determining module is configured to determine that an error occurs in the target data in the hard disk if the first CRC check value is inconsistent with the corresponding second CRC check value.
  • This application also provides a data verification device for a distributed storage system, the data verification device comprising: a memory and a processor; wherein the memory is used to store a computer program; the processor is used to execute the computer program to implement the following steps :
  • This application also provides a computer-readable storage medium with a computer program stored on the computer-readable storage medium, and when the computer program is executed by a processor, the following steps are implemented:
  • the computer-readable storage medium may include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk, etc., which can store program codes Medium.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Computer Security & Cryptography (AREA)
  • Quality & Reliability (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Debugging And Monitoring (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

La présente invention concerne un procédé de contrôle de données pour un système de stockage distribué, et un appareil associé. Le procédé comprend : la lecture de données cibles à partir d'un disque dur, et le calcul d'une valeur CRC de données de chaque longueur d'intervalle de contrôle prédéfinie dans les données cibles, et l'obtention d'une première valeur CRC ; lorsque les données cibles sont lues et écrites à partir d'une base de données, le calcul d'une deuxième valeur CRC obtenue à partir de la valeur CRC des données de chaque longueur d'intervalle de contrôle prédéfinie dans les données cibles ; la comparaison respectivement de chaque première valeur CRC à la deuxième valeur CRC correspondante ; et si la première valeur CRC est incohérente avec la deuxième valeur CRC correspondante, la détermination que les données cibles dans le disque dur présentent une erreur. Selon le procédé de contrôle de données, la position d'anomalie de données peut être identifiée avec précision, et un contrôle de données efficace peut être garanti.
PCT/CN2020/110952 2019-12-31 2020-08-25 Procédé de contrôle de données pour système de stockage distribué, et appareil associé WO2021135280A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911411044.2 2019-12-31
CN201911411044.2A CN111176885A (zh) 2019-12-31 2019-12-31 一种分布式存储系统的数据校验方法及相关装置

Publications (1)

Publication Number Publication Date
WO2021135280A1 true WO2021135280A1 (fr) 2021-07-08

Family

ID=70657748

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/110952 WO2021135280A1 (fr) 2019-12-31 2020-08-25 Procédé de contrôle de données pour système de stockage distribué, et appareil associé

Country Status (2)

Country Link
CN (1) CN111176885A (fr)
WO (1) WO2021135280A1 (fr)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111176885A (zh) * 2019-12-31 2020-05-19 浪潮电子信息产业股份有限公司 一种分布式存储系统的数据校验方法及相关装置
CN112416891B (zh) * 2020-11-26 2023-11-28 北京天融信网络安全技术有限公司 数据检测方法、装置、电子设备及可读存储介质
CN112667431A (zh) * 2020-12-23 2021-04-16 江西兴泰科技有限公司 一种电子价签波形数据的校对方法
CN113704150B (zh) * 2021-08-13 2023-08-04 苏州浪潮智能科技有限公司 一种用户态下dma数据缓存一致性方法、装置及系统
CN114579352A (zh) * 2022-04-29 2022-06-03 阿里云计算有限公司 数据重构方法以及装置
CN115729477A (zh) * 2023-01-09 2023-03-03 苏州浪潮智能科技有限公司 分布式存储io路径数据写入、读取方法、装置和设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0596340B1 (fr) * 1992-11-04 1999-03-03 Mitsubishi Denki Kabushiki Kaisha Circuit de correction d'erreur Reed-Solomon avec détection d'erreurs par CRC
US8266499B2 (en) * 2009-05-28 2012-09-11 Kabushiki Kaisha Toshiba CRC protection of data stored in XOR buffer
CN107888344A (zh) * 2016-09-29 2018-04-06 中兴通讯股份有限公司 一种误码检测的方法、设备和系统
US10205470B2 (en) * 2014-02-14 2019-02-12 Samsung Electronics Co., Ltd System and methods for low complexity list decoding of turbo codes and convolutional codes
CN109918226A (zh) * 2019-02-26 2019-06-21 平安科技（深圳）有限公司 一种静默错误检测方法、装置及存储介质
CN111176885A (zh) * 2019-12-31 2020-05-19 浪潮电子信息产业股份有限公司 一种分布式存储系统的数据校验方法及相关装置

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101452409B (zh) * 2007-12-04 2010-10-13 无锡江南计算技术研究所 数据校验冗余方法及装置
CN102184260B (zh) * 2011-06-09 2013-07-10 中国人民解放军国防科学技术大学 一种云计算环境下的海量数据存取方法
CN104978336A (zh) * 2014-04-08 2015-10-14 云南电力试验研究院（集团）有限公司电力研究院 基于Hadoop分布式计算平台的非结构化数据存储系统
CN108573007A (zh) * 2017-06-08 2018-09-25 北京金山云网络技术有限公司 检测数据一致性的方法、装置、电子设备及存储介质
CN107807792A (zh) * 2017-10-27 2018-03-16 郑州云海信息技术有限公司 一种基于副本存储系统的数据处理方法及相关装置
CN108875061A (zh) * 2018-06-29 2018-11-23 郑州云海信息技术有限公司 一种分布式文件系统的一致性测试方法及相关装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0596340B1 (fr) * 1992-11-04 1999-03-03 Mitsubishi Denki Kabushiki Kaisha Circuit de correction d'erreur Reed-Solomon avec détection d'erreurs par CRC
US8266499B2 (en) * 2009-05-28 2012-09-11 Kabushiki Kaisha Toshiba CRC protection of data stored in XOR buffer
US10205470B2 (en) * 2014-02-14 2019-02-12 Samsung Electronics Co., Ltd System and methods for low complexity list decoding of turbo codes and convolutional codes
CN107888344A (zh) * 2016-09-29 2018-04-06 中兴通讯股份有限公司 一种误码检测的方法、设备和系统
CN109918226A (zh) * 2019-02-26 2019-06-21 平安科技（深圳）有限公司 一种静默错误检测方法、装置及存储介质
CN111176885A (zh) * 2019-12-31 2020-05-19 浪潮电子信息产业股份有限公司 一种分布式存储系统的数据校验方法及相关装置

Also Published As

Publication number Publication date
CN111176885A (zh) 2020-05-19

Similar Documents

Publication Publication Date Title
WO2021135280A1 (fr) Procédé de contrôle de données pour système de stockage distribué, et appareil associé
US9003103B2 (en) Nonvolatile media dirty region tracking
EP2756399B1 (fr) Recherche et réparation de données
CN111078662B (zh) 一种区块链数据存储方法与装置
CN107479823B (zh) 随机读写文件测试中的数据校验方法和装置
US9329799B2 (en) Background checking for lost writes and data corruption
CN103778030B (zh) 日志子系统写入方法、错误追踪方法及处理器
CN107807792A (zh) 一种基于副本存储系统的数据处理方法及相关装置
US7827440B1 (en) Re-synchronizing corrupted data
BR112012031912B1 (pt) método implementado por computador, sistema e meio de armazenamento de computador
WO2017215377A1 (fr) Procédé et dispositif de traitement d'erreur de mémoire permanente
CN110008129B (zh) 一种存储定时快照的可靠性测试方法、装置及设备
WO2017143843A1 (fr) Procédé et dispositif de récupération de métadonnées
US10642508B2 (en) Method to limit impact of partial media failure of disk drive and detect/report the loss of data for objects due to partial failure of media
US9086990B2 (en) Bitline deletion
CN112558868A (zh) 一种配置数据存储的方法和装置及设备
CN105868127A (zh) 一种数据存储、读取方法及装置
KR20170031004A (ko) 데이터의 사일런트 커럽션을 감지하는 시스템들 및 그것의 동작 방법들
US7577804B2 (en) Detecting data integrity
CN113259410A (zh) 一种基于分布式存储的数据传输校验方法及系统
US9262264B2 (en) Error correction code seeding
WO2021056798A1 (fr) Dispositif, appareil et procédé de réparation de métadonnées et support de stockage
WO2018218814A1 (fr) Procédé de stockage d'enregistrement de transaction hors ligne, ainsi que dispositif informatique et support de stockage
US9152637B1 (en) Just-in time formatting of file system metadata
US11669413B2 (en) Object metadata maintenance pertinent to object restoration

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20910095

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20910095

Country of ref document: EP

Kind code of ref document: A1