公开(公告)号：US09489261B2

公开(公告)日：2016-11-08

申请号：US14968496

申请日：2015-12-14

Applicant: NetApp, Inc.

Inventor： Wei Sun ,  Donald Humlicek ,  Theresa Segura

IPC: G06F12/00 ,  G06F11/10 ,  G06F3/06

CPC classification number: G06F11/1076 ,  G06F3/0619 ,  G06F3/064 ,  G06F3/0689 ,  G06F11/108 ,  G06F2211/1057 ,  G06F2211/1059

Abstract: Technology is disclosed for performing background initialization on protection information enabled storage volumes or drives. In some embodiments, a storage controller generates multiple I/O requests for stripe segments of each drive (e.g., disk) of multiple drives of a RAID-based system (e.g., RAID-based disk array). The I/O requests are then sorted for each of the drives according to a pre-determined arrangement and initiated in parallel to the disks while enforcing the pre-determined arrangement. Sorting and issuing the I/O requests in the manner described herein can, for example, reduce drive head movement resulting in faster storage subsystem initialization.

公开(公告)号：US20160170833A1

公开(公告)日：2016-06-16

申请号：US14567743

申请日：2014-12-11

Applicant: NetApp, Inc.

Inventor： Theresa Segura ,  Wei Sun ,  Martin Parrish ,  Donald R. Humlicek

IPC: G06F11/10 ,  G06F11/20

CPC classification number: G06F11/1088 ,  G06F11/1471 ,  G06F11/1474 ,  G06F11/2058 ,  G06F11/2069 ,  G06F11/2089 ,  G06F11/2094 ,  G06F2211/1009

Abstract: A system and method for maintaining operation of a storage array with one or more failed storage devices and for quickly recovering when failing devices are replaced are provided. In some embodiments, the method includes receiving a data transaction directed to a volume and determining that a storage device associated with the volume is inoperable. In response to determining that the storage device is inoperable, a data extent is recorded in a change log in a storage controller cache. The data extent is associated with the data transaction and allocated to the storage device that is inoperable. The data transaction is performed using at least one other storage device associated with the volume, and data allocated to the storage device is subsequently reconstructed using the recorded data extent.

Abstract translation: 提供了一种用于维护具有一个或多个故障存储设备的存储阵列的操作并且用于在更换故障设备时快速恢复的系统和方法。 在一些实施例中，该方法包括接收指向卷的数据事务，并确定与该卷相关联的存储设备是不可操作的。 响应于确定存储设备不可操作，数据范围被记录在存储控制器高速缓存中的更改日志中。 数据范围与数据事务相关联，并分配给不可操作的存储设备。 使用与卷相关联的至少一个其他存储设备来执行数据事务，并且随后使用所记录的数据扩展来重构分配给存储设备的数据。

公开(公告)号：US20140325265A1

公开(公告)日：2014-10-30

申请号：US13956013

申请日：2013-07-31

Applicant: NetApp, Inc.

Inventor： Wei Sun ,  Donald Humlicek ,  Theresa Segura

IPC: G06F11/10

CPC classification number: G06F11/1076 ,  G06F3/0619 ,  G06F3/064 ,  G06F3/0689 ,  G06F11/108 ,  G06F2211/1057 ,  G06F2211/1059

Abstract: Technology is disclosed for performing background initialization on protection information enabled storage volumes or drives. In some embodiments, a storage controller generates multiple I/O requests for stripe segments of each drive (e.g., disk) of multiple drives of a RAID-based system (e.g., RAID-based disk array). The I/O requests are then sorted for each of the drives according to a pre-determined arrangement and initiated in parallel to the disks while enforcing the pre-determined arrangement. Sorting and issuing the I/O requests in the manner described herein can, for example, reduce drive head movement resulting in faster storage subsystem initialization.

Abstract translation: 公开了用于在启用保护信息的存储卷或驱动器上执行后台初始化的技术。 在一些实施例中，存储控制器为基于RAID的系统（例如，基于RAID的磁盘阵列）的多个驱动器的每个驱动器（例如，磁盘）的条带分段生成多个I / O请求。 然后根据预定的布置对每个驱动器对I / O请求进行排序，并且在执行预定的布置的同时并行地发送到磁盘。 以本文所描述的方式排序和发布I / O请求可以例如减少驱动头移动，从而导致更快的存储子系统初始化。

公开(公告)号：US20240289240A1

公开(公告)日：2024-08-29

申请号：US18652325

申请日：2024-05-01

Applicant: NetApp, Inc.

Inventor： Wei Sun ,  Anil Paul Thoppil ,  Anne Marie Vasu

IPC: G06F11/16 ,  G06F3/06 ,  G06F11/10 ,  G06F11/30 ,  G06F16/27

CPC classification number: G06F11/1662 ,  G06F3/0622 ,  G06F3/064 ,  G06F3/0679 ,  G06F11/1088 ,  G06F11/3034 ,  G06F16/27

Abstract: Systems and methods that make use of cluster-level redundancy within a distributed storage system to address various node-level error scenarios are provided. Rather than making use of a generalized one-size-fits-all approach in an effort to reduce complexity, an approach tailored to the node-level error scenario at issue may be performed to avoid doing more than necessary. According to one embodiment, responsive to identification of a failed RAID stripe by a node of a cluster of a distributed storage management system, for each block ID of multiple block IDs associated with the failed RAID stripe, a data block is restored corresponding to the block ID by reading the data block from another node of the cluster having a redundant copy of the data block; and writing the redundant copy of the data block to a storage area of the node that is unaffected by the failed RAID stripe.

公开(公告)号：US20230350850A1

公开(公告)日：2023-11-02

申请号：US17732046

申请日：2022-04-28

Applicant: NetApp Inc.

Inventor： Anil Paul Thoppil ,  Wei Sun ,  Meera Odugoudar ,  Szu-Wen Kuo ,  Santhosh Selvaraj

IPC: G06F3/06

CPC classification number: G06F3/0605 ,  G06F3/0685 ,  G06F3/0649

Abstract: Techniques are provided for implementing a defragmentation process during a merge operation performed by a re-compaction process upon a log structured merge tree. The log structured merge tree is used to store keys of key-value pairs within a key-value store. As the log structured merge tree fills with keys over time, the re-compaction process is performed to merge keys down to lower levels of the log structured merge tree to re-compact the keys. Re-compaction can result in fragmentation because there is a lack of spatial locality of where the re-compaction operations re-writes the keys within storage. Fragmentation increases read and write amplification when accessing the keys stored in different locations within the storage. Accordingly, the defragmentation process is performed during a last merge operation of the re-compaction process in order to store keys together within the storage, thus reducing read and write amplification when accessing the keys.

公开(公告)号：US11288186B2

公开(公告)日：2022-03-29

申请号：US16856228

申请日：2020-04-23

Applicant: NetApp, Inc.

Inventor： Alyssa Proulx ,  Wei Sun

IPC: G06F12/02 ,  G06F12/0864 ,  G06F11/30

Abstract: A system, method, and machine-readable storage medium for performing garbage collection in a distributed storage system are provided. In some embodiments, an efficiency level of a garbage collection process is monitored. The garbage collection process may include removal of one or more data blocks of a set of data blocks that is referenced by a set of content identifiers. The set of slice services and the set of data blocks may reside in a cluster, and a set of filters may indicate whether the set of data blocks is in-use. At least one parameter of a filter of the set of filters may be adjusted (e.g., increased or reduced) if the efficiency level is below the efficiency threshold. Garbage collection may be performed on the set of data blocks in accordance with the set of filters.

公开(公告)号：US20210334208A1

公开(公告)日：2021-10-28

申请号：US16856228

申请日：2020-04-23

Applicant: NetApp, Inc.

Inventor： Alyssa Proulx ,  Wei Sun

IPC: G06F12/02 ,  G06F12/0864 ,  G06F11/30

Abstract: A system, method, and machine-readable storage medium for performing garbage collection in a distributed storage system are provided. In some embodiments, an efficiency level of a garbage collection process is monitored. The garbage collection process may include removal of one or more data blocks of a set of data blocks that is referenced by a set of content identifiers. The set of slice services and the set of data blocks may reside in a cluster, and a set of filters may indicate whether the set of data blocks is in-use. At least one parameter of a filter of the set of filters may be adjusted (e.g., increased or reduced) if the efficiency level is below the efficiency threshold. Garbage collection may be performed on the set of data blocks in accordance with the set of filters.

公开(公告)号：US09830094B2

公开(公告)日：2017-11-28

申请号：US14746938

申请日：2015-06-23

Applicant: NetApp, Inc.

Inventor： Mahmoud K. Jibbe ,  Charles D. Binford ,  Wei Sun

IPC: G06F12/00 ,  G06F13/00 ,  G06F13/28 ,  G06F3/06 ,  G06F12/0875

CPC classification number: G06F3/0619 ,  G06F3/0617 ,  G06F3/0632 ,  G06F3/0637 ,  G06F3/0665 ,  G06F3/0689 ,  G06F12/0875 ,  G06F12/0891 ,  G06F2212/452

Abstract: A system, method, and computer program product is described for providing dynamic enabling and/or disabling of protection information (PI) in array systems during operation. A storage system receives a request to transition a volume from PI disabled to PI enabled during regular operation. The storage system synchronizes and purges the cache associated with the target volume. The storage system initiates an immediate availability format (IAF-PI) process to initialize PI for the associated data blocks of the volume's storage devices. The storage system continues receiving I/O requests as the IAF-PI process sweeps through the storage devices. The storage system inserts and checks PI for the write data as it is written to the storage devices. The storage system inserts PI for requested data above the IAF-PI boundary and checks PI for requested data below the IAF-PI boundary. The transition remains an online process that avoids downtime.

公开(公告)号：US20160098322A1

公开(公告)日：2016-04-07

申请号：US14968496

申请日：2015-12-14

Applicant: NetApp, Inc.

Inventor： Wei Sun ,  Donald Humlicek ,  Theresa Segura

IPC: G06F11/10 ,  G06F3/06

CPC classification number: G06F11/1076 ,  G06F3/0619 ,  G06F3/064 ,  G06F3/0689 ,  G06F11/108 ,  G06F2211/1057 ,  G06F2211/1059

Abstract: Technology is disclosed for performing background initialization on protection information enabled storage volumes or drives. In some embodiments, a storage controller generates multiple I/O requests for stripe segments of each drive (e.g., disk) of multiple drives of a RAID-based system (e.g., RAID-based disk array). The I/O requests are then sorted for each of the drives according to a pre-determined arrangement and initiated in parallel to the disks while enforcing the pre-determined arrangement. Sorting and issuing the I/O requests in the manner described herein can, for example, reduce drive head movement resulting in faster storage subsystem initialization.

Abstract translation: 公开了用于在启用保护信息的存储卷或驱动器上执行后台初始化的技术。 在一些实施例中，存储控制器为基于RAID的系统（例如，基于RAID的磁盘阵列）的多个驱动器的每个驱动器（例如，磁盘）的条带分段生成多个I / O请求。 然后根据预定的布置对每个驱动器对I / O请求进行排序，并且在执行预定的布置的同时并行地发送到磁盘。 以本文所描述的方式排序和发布I / O请求可以例如减少驱动头移动，从而导致更快的存储子系统初始化。

公开(公告)号：US20140325157A1

公开(公告)日：2014-10-30

申请号：US13871783

申请日：2013-04-26

Applicant: NetApp, Inc.

Inventor： Satish Sangapu ,  Wei Sun ,  Keith Moyer ,  Charles Binford

IPC: G06F3/06 ,  G06F12/08

CPC classification number: G06F3/061 ,  G06F3/0659 ,  G06F3/0689 ,  G06F11/2071 ,  G06F11/2082 ,  G06F12/0866 ,  G06F12/0875 ,  G06F2212/466

Abstract: An I/O processing stack includes a proxy that can provide processing services for access requests to initialized and uninitialized storage regions. For a write request, the proxy stores write information in a write metadata repository. If the write is requested for an address in an initialized storage region of the storage system, the proxy performs a write to the initialized region based on region information in the write I/O access request. If the write is requested for an address in an uninitialized storage region of the storage system, the proxy performs an on-demand initialization of the storage region and then performs a write to the storage region based on region information provided by the proxy.

Abstract translation: I / O处理堆栈包括可以为初始化和未初始化的存储区域的访问请求提供处理服务的代理。 对于写入请求，代理将写入信息存储在写入元数据存储库中。 如果对存储系统的初始化存储区域中的地址请求写入，则代理根据写入I / O访问请求中的区域信息执行对初始化区域的写入。 如果对存储系统的未初始化存储区域中的地址请求写入，则代理执行存储区域的按需初始化，然后基于由代理提供的区域信息对存储区域进行写入。