公开(公告)号：US11126591B2

公开(公告)日：2021-09-21

申请号：US16669574

申请日：2019-10-31

Applicant: Quantum Corporation

Inventor： Roderick B. Wideman ,  Don Doerner ,  Michael Zeis

IPC: G06F16/13 ,  G06F16/17 ,  G06F16/16

Abstract: Example apparatus and methods identify files that are so small or so large that they compromise the efficient operation of a file system that uses re-assignable one-to-one inodes and inode numbers. Small files are aggregated into collections of files and large files are subdivided into collections of smaller files. Information for locating multiple related files with fewer lookups is generated and stored in a folder. An inode having a new type of inode number is then created. The new type of inode number encodes information for finding the folder. The encoded information may include a folder identifier that acts as a primary key into a database that is configured to locate a member of the aggregated or subdivided files with a single lookup. A filter file system may be updated with the new inode. The new inode number is unique within the filter file system and may not be re-assigned.

公开(公告)号：US10168963B2

公开(公告)日：2019-01-01

申请号：US14944251

申请日：2015-11-18

Applicant: Quantum Corporation

Inventor： Don Doerner

IPC: G06F3/06 ,  G06F11/14

Abstract: Storage conditioning for a data storage system having D data storage devices (DSDs) is provided. E erasure codes (ECs) for an object are stored in the system, D>E. A map of d E-sized vectors of the D DSDs is produced. A DSD appears in e vectors. The ratio d/e is the reduced form of D/E. A hash value is produced for the object. A destination vector for storing the ECs is selected using the hash value according to a pre-determined, substantially uniform distribution. A compromised vector affected by a first DSD becoming unable to store ECs is identified. An intact vector that is not affected by the first DSD is identified. A complete set of ECs is produced from an incomplete set of ECs in the compromised vector and is distributed to the intact vector then copied back when the compromised vector is once again intact.

公开(公告)号：US09450617B2

公开(公告)日：2016-09-20

申请号：US14455529

申请日：2014-08-08

Applicant: Quantum Corporation

Inventor： Don Doerner

IPC: H03M13/37

CPC classification number: H03M13/3761 ,  H03M13/373

Abstract: Example apparatus and methods selectively replicate some erasure codes associated with a message and selectively distribute, without replicating, other erasure codes associated with the message. The message may have k symbols and n erasure codes may have been generated for the message, n>=k. In one embodiment, erasure codes that store plaintext information from the message (e.g., un-encoded symbols) may be replicated (e.g., sent to all devices using erasure codes associated with the message) while erasure codes that do not store plaintext information may be distributed (e.g., selectively moved to less than all devices) without being replicated. Some (e.g., less than k) erasure codes that do not store plaintext information may be stored unencrypted in the cloud. The generator matrix will not be stored in the cloud.

Abstract translation: 示例性设备和方法选择性地复制与消息相关联的一些擦除代码，并且选择性地分配与该消息相关联的其它擦除代码。 消息可以具有k个符号，并且可以为该消息生成n个擦除代码，n> = k。 在一个实施例中，可以复制来自消息（例如，未编码符号）的明文信息的擦除代码（例如，使用与消息相关联的擦除代码发送到所有设备），而不存储明文信息的擦除代码可以是 分布式（例如，选择性地移动到小于所有设备）而不被复制。 不存储明文信息的一些（例如小于k个）擦除代码可以在云中未被加密地存储。 发电机矩阵不会存储在云中。

公开(公告)号：US09431054B1

公开(公告)日：2016-08-30

申请号：US14727893

申请日：2015-06-02

Applicant: Quantum Corporation

Inventor： Turguy Goker ,  Suayb S. Arslan ,  Don Doerner ,  Hoa Le

IPC: G11B5/09 ,  G11B20/18 ,  G11B5/008 ,  G06F11/10 ,  H03M13/15 ,  H03M13/27 ,  H04L1/00 ,  H03M13/11 ,  G11B20/14

CPC classification number: G11B20/1833 ,  G06F11/1076 ,  G11B20/1426 ,  H03M13/116 ,  H03M13/154 ,  H03M13/2792 ,  H04L1/0057

Abstract: Methods, apparatus, and other embodiments associated with doubly distributing erasure encoded data in a data storage system are described. One example apparatus includes a set of data storage devices and a set of logics that includes an encoding logic that generates an erasure encoded object that includes code-words, and chunks the code-words into code-word chunks, and a distribution logic that interleaves members of the set of code-word chunks into a plurality of records, and distributes the records across the data storage devices and within individual data storage devices. Example apparatus may include a read logic that reads the plurality of stored records from the data storage devices, and ignores read errors, and a repair logic that monitors the set of data storage devices, replaces or repairs failing data storage devices, generates replacement records, and stores the replacement records on a replacement data storage device.

Abstract translation: 描述了与在数据存储系统中双重分配擦除编码数据相关联的方法，装置和其他实施例。 一个示例性设备包括一组数据存储设备和一组逻辑，其包括生成包括码字的擦除编码对象的编码逻辑，并且将码字块编码为码字块，以及交织逻辑 将该组代码字块的成员分成多个记录，并且在数据存储设备和各个数据存储设备内分发记录。 示例性设备可以包括从数据存储设备读取多个存储的记录并忽略读取错误的读取逻辑，以及监视数据存储设备的集合，替换或修复故障数据存储设备的修复逻辑，生成替换记录， 并将替换记录存储在替换数据存储装置上。

公开(公告)号：US20160239380A1

公开(公告)日：2016-08-18

申请号：US14622976

申请日：2015-02-16

Applicant: Quantum Corporation

Inventor： Roderick Wideman ,  Don Doerner

IPC: G06F11/10 ,  G06F12/02 ,  H03M13/15 ,  G11C29/52

CPC classification number: G06F12/0246 ,  G06F2212/1044 ,  G06F2212/214 ,  G06F2212/7205 ,  H03M13/373 ,  H03M13/3761 ,  Y02D10/13

Abstract: Example apparatus and methods provide improved reclamation, garbage collection (GC) and defragmentation (defrag) for data storage devices including solid state drives (SSD) or shingled magnetic recording (SMR) drives. An erasure code (EC) layer that facilitates logically or physically erasing data from the SSD or SMR as a comprehensive GC or defrag is added to the SSD or SMR. Erased data may be selectively recreated from the EC layer as needed. Pre-planned EC write zones may be established to further optimize GC and defrag. Recreated data may be written to selected locations to further optimize SSD and SMR performance. Erasure code data may be distributed to co-operating devices to further improve GC or defrag. Example apparatus and methods may also facilitate writing data to an SMR drive using tape or VTL applications or processes and providing a pseudo virtual tape library on the SMR drive.

Abstract translation: 示例性设备和方法为包括固态驱动器（SSD）或带状磁记录（SMR）驱动器的数据存储设备提供改进的回收，垃圾回收（GC）和碎片整理（碎片整理）。 有助于将SSD或SMR的数据逻辑或物理擦除为全面GC或碎片整理的擦除代码（EC）层被添加到SSD或SMR。 可以根据需要从EC层选择性地重新创建被擦除的数据。 可以建立预先计划的EC写入区域，以进一步优化GC和碎片整理。 可以将已处理的数据写入选定位置，以进一步优化SSD和SMR性能。 擦除代码数据可以分发到协作设备以进一步改进GC或碎片整理。 示例性设备和方法还可以方便地使用磁带或VTL应用程序向SMR驱动器写入数据，或者在SMR驱动器上提供伪虚拟磁带库。

公开(公告)号：US20160043738A1

公开(公告)日：2016-02-11

申请号：US14455529

申请日：2014-08-08

Applicant: Quantum Corporation

Inventor： Don Doerner

IPC: H03M13/15

CPC classification number: H03M13/3761 ,  H03M13/373

Abstract: Example apparatus and methods selectively replicate some erasure codes associated with a message and selectively distribute, without replicating, other erasure codes associated with the message. The message may have k symbols and n erasure codes may have been generated for the message, n>=k. In one embodiment, erasure codes that store plaintext information from the message (e.g., un-encoded symbols) may be replicated (e.g., sent to all devices using erasure codes associated with the message) while erasure codes that do not store plaintext information may be distributed (e.g., selectively moved to less than all devices) without being replicated. Some (e.g., less than k) erasure codes that do not store plaintext information may be stored unencrypted in the cloud. The generator matrix will not be stored in the cloud.

Abstract translation: 示例性设备和方法选择性地复制与消息相关联的一些擦除代码，并且选择性地分配与该消息相关联的其它擦除代码。 消息可以具有k个符号，并且可以为该消息生成n个擦除代码，n> = k。 在一个实施例中，可以复制来自消息（例如，未编码符号）的明文信息的擦除代码（例如，使用与消息相关联的擦除代码发送到所有设备），而不存储明文信息的擦除代码可以是 分布式（例如，选择性地移动到小于所有设备）而不被复制。 不存储明文信息的一些（例如小于k个）擦除代码可以在云中未被加密地存储。 发电机矩阵不会存储在云中。

公开(公告)号：US20150378820A1

公开(公告)日：2015-12-31

申请号：US14314977

申请日：2014-06-25

Applicant: Quantum Corporation

Inventor： Don Doerner

IPC: G06F11/10

CPC classification number: G06F11/1076 ,  H03M13/373 ,  H03M13/3761

Abstract: Example apparatus and methods treat some erasure codes differently than other erasure codes. For example, erasure codes that are only involved in error-recovery may never be read and thus may be stored using a different approach than erasure codes that are involved in more regular data reading. If different types of data stores are available, then the erasure codes that are more likely to be read may be stored in data stores having a first (e.g., higher, faster) type of read performance while the erasure codes that are less likely to be read may be stored in data stores having a second (e.g., lower, slower, less expensive) type of read performance. Different data stores may be located on different data storage devices. Different data stores may even be located on a single data storage device.

Abstract translation: 示例性设备和方法与其他擦除代码不同的是对待一些擦除代码。 例如，仅涉及错误恢复的擦除代码可能永远不会被读取，并且因此可以使用与更常规数据读取中涉及的擦除代码不同的方法来存储。 如果不同类型的数据存储可用，那么更可能被读取的擦除代码可以被存储在具有第一（例如更高，更快）类型的读取性能的数据存储中，而不太可能读取的擦除代码 读取可以存储在具有第二（例如，较低，较慢，较便宜）的读取性能的数据存储中。 不同的数据存储可以位于不同的数据存储设备上。 不同的数据存储器甚至可以位于单个数据存储设备上。

公开(公告)号：US10761758B2

公开(公告)日：2020-09-01

申请号：US15386484

申请日：2016-12-21

Applicant: Quantum Corporation

Inventor： Don Doerner ,  Michael Zeis ,  Roderick Wideman ,  Greg Wade ,  Steve Lord ,  Turguy Goker

IPC: G06F16/22 ,  G06F3/06 ,  G06F12/02 ,  G06F16/13 ,  G06F16/174

Abstract: Embodiments include a data aware deduplicating object store. The data aware deduplicating data store includes a consistent hashing logic that manages a consistent hashing architecture for the object store. The consistent hashing architecture includes a metadata ring and a bulk ring. The consistent hashing architecture may be a multiple ring architecture comprising a metadata ring and two or more bulk rings. A bulk ring may include a key/value (k/v) data store, where a k/v data store stores a shard of an index and a reference count that facilitates the individual approach to garbage collection or data reclamation. The data aware deduplicating data store also includes a deduplication logic that provides data deduplication for data to be stored in the object store. The deduplication logic performs variable length deduplication and provides a shared nothing approach.

公开(公告)号：US20190243766A1

公开(公告)日：2019-08-08

申请号：US16268868

申请日：2019-02-06

Applicant: Quantum Corporation

Inventor： Don Doerner

IPC: G06F12/0862

CPC classification number: G06F12/0862 ,  G06F2212/1021 ,  G06F2212/6024

Abstract: Adaptive pre-fetching devices can predict data placement to improve the operating and/or electrical efficiency of a data storage system. A future input/output operation can be predicted from a current input/output operation, the state of the data storage apparatus, relationships between data currently being processed and data previously processed, or other factors. The apparatus and methods can improve data storage efficiency by selectively pre-fetching data, relocating data on the data storage apparatus, the backing storage, or within a plurality of data storage apparatus based on working set predictors to reduce cache misses or outperform fetch processes from the backing storage.

公开(公告)号：US20180260583A1

公开(公告)日：2018-09-13

申请号：US15632584

申请日：2017-06-26

Applicant: Quantum Corporation

Inventor： Don Doerner ,  Gregory L. Wade

IPC: G06F21/64 ,  G06F21/60 ,  H04L9/06

Abstract: Example methods and apparatus asynchronously verify data stored in a cloud data storage system. One embodiment comprises a monitoring circuit that determines if a data auditing condition associated with a cloud storage system or archived data stored in the cloud storage system has been met, a metadata mirror circuit that controls a metadata mirror to provide metadata, including a first checksum, associated with the archived data to the apparatus, a checksum circuit that computes a second checksum based on the archived data, a verification circuit that generates an audit of the first checksum and the second checksum by comparing the second checksum with the first checksum, and a reporting circuit that generates a log of the audit, that provides the log to the data storage system, and that provides a notification of a data integrity failure to a user associated with the archived data.