公开(公告)号：US06959313B2

公开(公告)日：2005-10-25

申请号：US10616128

申请日：2003-07-08

Applicant: Vikram Kapoor ,  Kurt Alan Shoens ,  Mark Steven Schultz ,  Rex Rilen Hamilton

Inventor： Vikram Kapoor ,  Kurt Alan Shoens ,  Mark Steven Schultz ,  Rex Rilen Hamilton

IPC: G06F20060101 ,  G06F12/00 ,  G06F17/30

CPC classification number: G06F17/30067 ,  Y10S707/915 ,  Y10S707/99953 ,  Y10S707/99956

Abstract: The present invention relates to methods and systems of snapshot management of a file system in a data storage system. To represent the snapshots, the invention maintains pointers to the root block pointer of each snapshot. When the active file system is modified, this invention avoids overwriting any blocks used by previous snapshots by allocating new blocks for the modified blocks. When the invention needs to put an established block in a new location, it must update a parent block to point to the new location. The update to the parent block may then require allocating a new block for the new parent block and so forth. Parts of the file system not modified since a snapshot remain in place. The amount of space required to represent snapshots scales with the fraction of the file system that users modify. To maintain snapshot integrity, this invention keeps track of the first and last snapshots that use each block in space map blocks spread throughout the file system data space. When users delete snapshots, this invention may use a background process to find blocks no longer used by any snapshot and makes them available for future use.

公开(公告)号：US07380059B2

公开(公告)日：2008-05-27

申请号：US11408209

申请日：2006-04-19

Applicant: David Alan Burton

Inventor： David Alan Burton

IPC: G06F12/00

CPC classification number: G06F12/0862 ,  G06F12/0871 ,  G06F12/123 ,  G06F2201/84 ,  G06F2212/284 ,  G06F2212/285 ,  G06F2212/311

Abstract: The present invention relates to a cache memory management system suitable for use with snapshot applications. The system includes a cache directory including a hash table, hash table elements, cache line descriptors, and cache line functional pointers, and a cache manager running a hashing function that converts a request for data from an application to an index to a first hash table pointer in the hash table. The first hash table pointer in turn points to a first hash table element in a linked list of hash table elements where one of the hash table elements of the linked list of hash table elements points to a first cache line descriptor in the cache directory and a cache memory including a plurality of cache lines, wherein the first cache line descriptor has a one-to-one association with a first cache line. The present invention also provides for a method converting a request for data to an input to a hashing function, addressing a hash table based on a first index output from the hashing function, searching the hash table elements pointed to by the first index for the requested data, determining the requested data is not in cache memory, and allocating a first hash table element and a first cache line descriptor that associates with a first cache line in the cache memory.

Abstract translation: 本发明涉及适用于快照应用的缓存存储器管理系统。 该系统包括缓存目录，其包括哈希表，哈希表元素，高速缓存行描述符和高速缓存行功能指针，以及运行散列函数的高速缓存管理器，该数据从数据从应用程序转换为索引到第一散列表 指针在哈希表中。 第一哈希表指针又指向哈希表元素的链表中的第一哈希表元素，其中哈希表元素的链表的哈希表元素之一指向高速缓存目录中的第一高速缓存行描述符， 包括多个高速缓存行的高速缓冲存储器，其中第一高速缓存行描述符与第一高速缓存行具有一对一关联。 本发明还提供了一种将数据请求转换为输入到散列函数的方法，基于从散列函数输出的第一索引寻址哈希表，搜索由第一索引指向的哈希表元素中所请求的哈希表元素 确定所请求的数据的数据不在高速缓冲存储器中，并且分配与缓存存储器中的第一高速缓存行相关联的第一散列表元素和第一高速缓存行描述符。

公开(公告)号：US07756844B2

公开(公告)日：2010-07-13

申请号：US12154494

申请日：2008-05-23

Applicant: Kurt Alan Shoens ,  Tod Kevin Johnson

Inventor： Kurt Alan Shoens ,  Tod Kevin Johnson

IPC: G06F17/30

CPC classification number: G06F17/3023 ,  G06F11/1435 ,  G06F2201/84

Abstract: The invention relates to a method of determining if a block was modified in a file system by comparing the versions of the base snapshot, the delta snapshot, and the space map block entry (b, e). In another aspect, the invention relates to a method of searching for blocks modified in a tree structured file system. The invention relates to methods and systems of snapshot management of a file system in a data storage system. To represent the snapshots, the invention maintains pointers to the root block pointer of each snapshot. When the active file system is modified, this invention avoids overwriting any blocks used by previous snapshots by allocating new blocks for the modified blocks. When the invention needs to put an established block in a new location, it must update a parent block to point to the new location. The update to the parent block may then require allocating a new block for the new parent block and so forth. Parts of the file system not modified since a snapshot remain in place. The amount of space required to represent snapshots scales with the fraction of the file system that users modify. To maintain snapshot integrity, this invention keeps track of the first and last snapshots that use each block in space map blocks spread throughout the file system data space. When users delete snapshots, this invention may use a background process to find blocks no longer used by any snapshot and makes them available for future use.

Abstract translation: 本发明涉及通过比较基本快照，增量快照和空间地图块条目（b，e）的版本来确定文件系统中块是否被修改的方法。 在另一方面，本发明涉及一种搜索在树结构化文件系统中修改的块的方法。 本发明涉及数据存储系统中文件系统的快照管理方法和系统。 为了表示快照，本发明保持指向每个快照的根块指针的指针。 当活动文件系统被修改时，本发明通过为修改的块分配新的块来避免覆盖先前快照所使用的任何块。 当本发明需要将一个已建立的块放在新位置时，它必须更新一个父块以指向新的位置。 然后，对父块的更新可能需要为新的父块等分配新块。 文件系统的部分因为快照保留到位而未被修改。 代表快照所需的空间量会随着用户修改的文件系统的一小部分而缩放。 为了保持快照完整性，本发明记录使用遍及文件系统数据空间的空间图块中的每个块的第一个和最后一个快照。 当用户删除快照时，本发明可以使用后台进程来查找不再被任何快照使用的块，并使之可供将来使用。

公开(公告)号：US07653669B2

公开(公告)日：2010-01-26

申请号：US11879230

申请日：2007-07-16

Applicant: Vikram Kapoor ,  Kurt Alan Shoens ,  Mark Steven Schultz ,  Rex Rilen Hamilton

Inventor： Vikram Kapoor ,  Kurt Alan Shoens ,  Mark Steven Schultz ,  Rex Rilen Hamilton

IPC: G06F17/30

CPC classification number: G06F17/30067 ,  Y10S707/915 ,  Y10S707/99953 ,  Y10S707/99956

Abstract: The present invention relates to methods and systems of snapshot management of a file system in a data storage system. To represent the snapshots, the invention maintains pointers to the root block pointer of each snapshot. When the active file system is modified, this invention avoids overwriting any blocks used by previous snapshots by allocating new blocks for the modified blocks. When the invention needs to put an established block in a new location, it must update a parent block to point to the new location. The update to the parent block may then require allocating a new block for the new parent block and so forth. Parts of the file system not modified since a snapshot remain in place. The amount of space required to represent snapshots scales with the fraction of the file system that users modify. To maintain snapshot integrity, this invention keeps track of the first and last snapshots that use each block in space map blocks spread throughout the file system data space. When users delete snapshots, this invention may use a background process to find blocks no longer used by any snapshot and makes them available for future use.

Abstract translation: 本发明涉及数据存储系统中文件系统的快照管理方法和系统。 为了表示快照，本发明保持指向每个快照的根块指针的指针。 当活动文件系统被修改时，本发明通过为修改的块分配新的块来避免覆盖先前快照所使用的任何块。 当本发明需要将一个已建立的块放在新位置时，它必须更新一个父块以指向新的位置。 然后，对父块的更新可能需要为新的父块等分配新块。 文件系统的部分因为快照保留到位而未被修改。 代表快照所需的空间量会随着用户修改的文件系统的一小部分而缩放。 为了保持快照完整性，本发明记录使用遍及文件系统数据空间的空间图块中的每个块的第一个和最后一个快照。 当用户删除快照时，本发明可以使用后台进程来查找不再被任何快照使用的块，并使之可供将来使用。

公开(公告)号：US07594044B2

公开(公告)日：2009-09-22

申请号：US11897431

申请日：2007-08-29

Applicant: Michael Allan Brewer ,  David Alan Burton ,  Michael Lee Workman

Inventor： Michael Allan Brewer ,  David Alan Burton ,  Michael Lee Workman

IPC: G06F3/00 ,  G06F12/00

CPC classification number: G06F3/0659 ,  G06F3/0605 ,  G06F3/0613 ,  G06F3/0631 ,  G06F3/0635 ,  G06F3/067

Abstract: The invention classifies volumes (e.g., file systems or LUNs) of a data storage system according to application requirements and allocates space for the volumes on storage devices (e.g., hard disk drives) accordingly. A person such as an IT administrator configures the volumes specifying size, type (e.g., file system or SAN LUN), and priority (e.g., high, medium, low, or archive). The host schedules I/O requests to the storage devices in priority queues using the volume definition to match the application requirements and reduce storage seek time between volumes of different priorities. The host also allocates high performance bands of the storage devices to high performance applications and lower performance bands to lower performance applications. In this manner, the data storage system places data on the band of the storage device that best supports its performance needs.

Abstract translation: 本发明根据应用需求对数据存储系统的卷（例如，文件系统或LUN）进行分类，并相应地为存储设备（例如，硬盘驱动器）上的卷分配空间。 诸如IT管理员之类的人员配置指定大小，类型（例如，文件系统或SAN LUN）以及优先级（例如，高，中，低或归档）的卷。 主机使用卷定义将I / O请求调度到优先级队列中的存储设备，以匹配应用程序需求，并减少不同优先级卷之间的存储寻道时间。 主机还将高性能频段的存储设备分配到高性能应用程序和较低性能频段，以降低性能应用程序。 以这种方式，数据存储系统将数据放置在最能支持其性能需求的存储设备的频带上。

公开(公告)号：US07418531B2

公开(公告)日：2008-08-26

申请号：US11122495

申请日：2005-05-04

Applicant: Michael Allan Brewer ,  David Alan Burton ,  Michael Lee Workman

Inventor： Michael Allan Brewer ,  David Alan Burton ,  Michael Lee Workman

IPC: G06F3/00 ,  G06F12/00

CPC classification number: G06F3/0659 ,  G06F3/0605 ,  G06F3/0613 ,  G06F3/0631 ,  G06F3/0635 ,  G06F3/067

Abstract: The invention classifies volumes (e.g., file systems or LUNs) of a data storage system according to application requirements and allocates space for the volumes on storage devices (e.g., hard disk drives) accordingly. A person such as an IT administrator configures the volumes specifying size, type (e.g., file system or SAN LUN), and priority (e.g., high, medium, low, or archive). The host schedules I/O requests to the storage devices in priority queues using the volume definition to match the application requirements and reduce storage seek time between volumes of different priorities. The host also allocates high performance bands of the storage devices to high performance applications and lower performance bands to lower performance applications. In this manner, the data storage system places data on the band of the storage device that best supports its performance needs.

Abstract translation: 本发明根据应用需求对数据存储系统的卷（例如，文件系统或LUN）进行分类，并相应地为存储设备（例如，硬盘驱动器）上的卷分配空间。 诸如IT管理员之类的人员配置指定大小，类型（例如，文件系统或SAN LUN）以及优先级（例如，高，中，低或归档）的卷。 主机使用卷定义将I / O请求调度到优先级队列中的存储设备，以匹配应用程序需求，并减少不同优先级卷之间的存储寻道时间。 主机还将高性能频段的存储设备分配到高性能应用程序和较低性能频段，以降低性能应用程序。 以这种方式，数据存储系统将数据放置在最能支持其性能需求的存储设备的频带上。

公开(公告)号：US07721022B2

公开(公告)日：2010-05-18

申请号：US12459067

申请日：2009-06-26

Applicant: Michael Allan Brewer ,  David Alan Burton ,  Michael Lee Workman

Inventor： Michael Allan Brewer ,  David Alan Burton ,  Michael Lee Workman

IPC: G06F3/00 ,  G06F12/00

CPC classification number: G06F3/0659 ,  G06F3/0605 ,  G06F3/0613 ,  G06F3/0631 ,  G06F3/0635 ,  G06F3/067

Abstract: The invention classifies volumes (e.g., file systems or LUNs) of a data storage system according to application requirements and allocates space for the volumes on storage devices (e.g., hard disk drives) accordingly. A person such as an IT administrator configures the volumes specifying size, type (e.g., file system or SAN LUN), and priority (e.g., high, medium, low, or archive). The host schedules I/O requests to the storage devices in priority queues using the volume definition to match the application requirements and reduce storage seek time between volumes of different priorities. The host also allocates high performance bands of the storage devices to high performance applications and lower performance bands to lower performance applications. In this manner, the data storage system places data on the band of the storage device that best supports its performance needs.

Abstract translation: 本发明根据应用需求对数据存储系统的卷（例如，文件系统或LUN）进行分类，并相应地为存储设备（例如，硬盘驱动器）上的卷分配空间。 诸如IT管理员之类的人员配置指定大小，类型（例如，文件系统或SAN LUN）以及优先级（例如，高，中，低或归档）的卷。 主机使用卷定义将I / O请求调度到优先级队列中的存储设备，以匹配应用程序需求，并减少不同优先级卷之间的存储寻道时间。 主机还将高性能频段的存储设备分配到高性能应用程序和较低性能频段，以降低性能应用程序。 以这种方式，数据存储系统将数据放置在最能支持其性能需求的存储设备的频带上。

公开(公告)号：US07496725B2

公开(公告)日：2009-02-24

申请号：US11598868

申请日：2006-11-13

Applicant: David Alan Burton ,  Noel Simen Otterness

Inventor： David Alan Burton ,  Noel Simen Otterness

IPC: G06F12/16 ,  G06F13/00

CPC classification number: G06F3/065 ,  G06F3/0608 ,  G06F3/0647 ,  G06F11/1402 ,  G06F2201/82

Abstract: The invention relates to methods of snapshot operation for a data storage system with a host communicating with a cache memory, a source Virtual Logical Unit Number (VLUN) and a target VLUN, including generating first metadata (e.g., bitmaps and log files pointers) to locate first snapshot data and to indicate when the first snapshot data is in the target VLUN and generating second metadata to locate second snapshot data and to indicate when the second snapshot data is in the target VLUN, wherein the first and second metadata locate the same data in the target VLUN. The invention relates to systems that implement the methods. It relates to destaging data to maintain data consistency including reading metadata for snapshots, searching the metadata to identify snapshots that require original data to be destaged, destaging the original data to target storage, and updating metadata to locate original data and indicate destage completion.

Abstract translation: 本发明涉及一种具有与高速缓冲存储器通信的主机的数据存储系统的快照操作方法，源虚拟逻辑单元号（VLUN）和目标VLUN，包括生成第一元数据（例如，位图和日志文件指针）到 定位第一快照数据并且指示第一快照数据何时在目标VLUN中并且生成第二元数据以定位第二快照数据并且指示何时第二快照数据在目标VLUN中，其中第一和第二元数据定位相同的数据 在目标VLUN中。 本发明涉及实现该方法的系统。 它涉及到数据的维护，以保持数据的一致性，包括读取快照的元数据，搜索元数据来识别需要去除原始数据的快照，将原始数据降级到目标存储，以及更新元数据以定位原始数据并指示流水完成。

公开(公告)号：US07380157B2

公开(公告)日：2008-05-27

申请号：US10975290

申请日：2004-10-27

Applicant: Michael Allan Brewer ,  Robert Arthur Ellis

Inventor： Michael Allan Brewer ,  Robert Arthur Ellis

IPC: G06F11/00

CPC classification number: G06F11/1076 ,  G06F11/2061 ,  G06F11/2094 ,  G06F2211/1045 ,  H04Q2213/13103

Abstract: The present invention relates to write operations in data storage systems. In an embodiment, the invention relates to a method of writing data across a plurality of disk arrays, including writing a first copy of the data as a first plurality of stripes corresponding to the plurality of disk arrays and writing a second copy of the data as a second plurality of stripes corresponding to the plurality of disk arrays, wherein the second copy of the data is staggered with respect to the first copy on the plurality of disk arrays. In another embodiment, the invention relates to a data storage system including a plurality of disk arrays and a host, coupled to the plurality of disk arrays, with a program that instructs the host to write a first copy of the data as a first plurality of stripes that correspond to the plurality of disk arrays and to write a second copy of the data as a second plurality of stripes that correspond to the plurality of disk arrays, wherein the second copy of the data is staggered with respect to the first copy on the plurality of disk arrays.

Abstract translation: 本发明涉及数据存储系统中的写入操作。 在一个实施例中，本发明涉及一种在多个磁盘阵列之间写入数据的方法，包括将数据的第一副本作为与多个磁盘阵列相对应的第一多个条带写入，并将数据的第二副本写入 对应于所述多个盘阵列的第二多个条带，其中所述数据的第二副本相对于所述多个盘阵列上的所述第一拷贝交错布置。 在另一个实施例中，本发明涉及一种数据存储系统，其包括耦合到多个磁盘阵列的多个磁盘阵列和主机，该程序指示主机写入数据的第一副本作为第一多个 对应于多个磁盘阵列的条带，并写入数据的第二副本作为对应于多个磁盘阵列的第二多个条带，其中数据的第二副本相对于第一副本交错 多个磁盘阵列。

公开(公告)号：US07379954B2

公开(公告)日：2008-05-27

申请号：US11407491

申请日：2006-04-19

Applicant: Kurt Alan Shoens

Inventor： Kurt Alan Shoens

IPC: G06F17/30

CPC classification number: G06F17/30088 ,  Y10S707/99953 ,  Y10S707/99956

Abstract: The present invention relates to methods and systems of snapshot management of a file system in a data storage system. To represent the snapshots, the invention maintains pointers to the root block pointer of each snapshot. When the active file system is modified, this invention avoids overwriting any blocks used by previous snapshots by allocating new blocks for the modified blocks. When the invention needs to put an established block in a new location, it must update a parent block to point to the new location. The update to the parent block may then require allocating a new block for the new parent block and so forth. Parts of the file system not modified since a snapshot remain in place. The amount of space required to represent snapshots scales with the fraction of the file system that users modify. To maintain snapshot integrity, this invention keeps track of the first and last snapshots that use each block in space map blocks spread throughout the file system data space. When users delete snapshots, this invention may use a background process to find blocks no longer used by any snapshot and makes them available for future use.

Abstract translation: 本发明涉及数据存储系统中文件系统的快照管理方法和系统。 为了表示快照，本发明保持指向每个快照的根块指针的指针。 当活动文件系统被修改时，本发明通过为修改的块分配新的块来避免覆盖先前快照所使用的任何块。 当本发明需要将一个已建立的块放在新位置时，它必须更新一个父块以指向新的位置。 然后，对父块的更新可能需要为新的父块等分配新块。 文件系统的部分因为快照保留到位而未被修改。 代表快照所需的空间量会随着用户修改的文件系统的一小部分而缩放。 为了保持快照完整性，本发明记录使用遍及文件系统数据空间的空间图块中的每个块的第一个和最后一个快照。 当用户删除快照时，本发明可以使用后台进程来查找不再被任何快照使用的块，并使之可供将来使用。