公开(公告)号：US20170262465A1

公开(公告)日：2017-09-14

申请号：US15068988

申请日：2016-03-14

Applicant: VMware, Inc.

Inventor： Edward J. Goggin ,  Sandeep Rangaswamy ,  Kiran Joshi ,  Srinath Premachandran ,  Gopala Suryanarayana ,  Radu Berinde ,  Wenguang Wang ,  Jorge Guerra Delgado ,  Jin Zhang ,  Radhika Vullikanti

IPC: G06F17/30 ,  H04L29/08

CPC classification number: G06F16/1748 ,  H04L67/1004 ,  H04L67/1097

Abstract: In accordance with the present disclosure, files may be deduplicated in a distributed storage system having a plurality of storage volumes. A uniqueness metric for each file may indicate a degree of deduplication of the respective data files in the given storage volume. The uniqueness metric may be used to identify files for rebalancing in the distributed storage system. The uniqueness metric may be efficiently calculated with enough accuracy using a sampling methodology.

公开(公告)号：US10168942B2

公开(公告)日：2019-01-01

申请号：US15255063

申请日：2016-09-01

Applicant: VMware, Inc.

Inventor： Eric Knauft ,  Edward Joseph Goggin ,  Ruocheng Li ,  Radhika Vullikanti ,  Gopala Suryanarayana

IPC: G06F3/06 ,  G06F11/00

Abstract: Exemplary methods, apparatuses, and systems determine that a disk drive in a distributed storage system is failing. In response to determining that the disk drive is failing, it is determined if an object has object components stored on the failing disk drive. Object components are flagged based on whether the object components will become inaccessible in the event the disk drive fails or is unmounted. Object components that are flagged with an evacuate flag are migrated from the failing disk drive to one or more other disk drives in the distributed storage system to prevent objects from becoming inaccessible.

公开(公告)号：US11182344B2

公开(公告)日：2021-11-23

申请号：US15068988

申请日：2016-03-14

Applicant: VMware, Inc.

Inventor： Edward J Goggin ,  Sandeep Rangaswamy ,  Kiran Joshi ,  Srinath Premachandran ,  Gopala Suryanarayana ,  Radu Berinde ,  Wenguang Wang ,  Jorge Guerra Delgado ,  Jin Zhang ,  Radhika Vullikanti

IPC: G06F16/00 ,  G06F16/174 ,  H04L29/08

Abstract: In accordance with the present disclosure, files may be deduplicated in a distributed storage system having a plurality of storage volumes. A uniqueness metric for each file may indicate a degree of deduplication of the respective data files in the given storage volume. The uniqueness metric may be used to identify files for rebalancing in the distributed storage system. The uniqueness metric may be efficiently calculated with enough accuracy using a sampling methodology.

公开(公告)号：US10346063B2

公开(公告)日：2019-07-09

申请号：US15356413

申请日：2016-11-18

Applicant: VMware, Inc.

Inventor： Jin Zhang ,  Radhika Vullikanti ,  Gopala Suryanarayana ,  Varun Shah

IPC: G06F11/00 ,  G06F3/06 ,  G06F11/18 ,  G06F11/20

Abstract: Exemplary methods, apparatuses, and systems determine that quorum can be maintained for a storage object in a distributed storage system in the event a defined maximum number of first partitions in a first level of storage and a second partitions in a second level of storage fail. When it is determined that there are insufficient numbers of first partitions and/or second partitions, additional first partitions and/or second partitions are associated with the storage object in the distributed storage system. A number of votes is calculated for distribution and an allocation is defined for assigning the votes to each component and witness component of the storage object.

公开(公告)号：US10895995B2

公开(公告)日：2021-01-19

申请号：US16431648

申请日：2019-06-04

Applicant: VMware, Inc.

Inventor： Jorge Guerra Delgado ,  Jin Zhang ,  Radhika Vullikanti ,  Abhishek Gupta

IPC: G06F3/06 ,  G06F16/23 ,  G06F16/174 ,  G06F9/455 ,  G06F16/25 ,  H04L29/08

Abstract: A logical group of data blocks stored in a first node is migrated to a second node according to a method that includes determining a first metric for each logical group of data blocks stored in the first node, the first metric representing a total size of the data blocks in the logical group, determining a second metric for each logical group of data blocks stored in the first node, the second metric representing a total size of the data blocks in the logical group that are uniquely stored in the first node, and selecting a logical group of data blocks for migration from the first node to the second node based on the first metric and the second metric.

公开(公告)号：US10359966B2

公开(公告)日：2019-07-23

申请号：US15592834

申请日：2017-05-11

Applicant: VMware, Inc.

Inventor： Jorge Guerra Delgado ,  Jin Zhang ,  Radhika Vullikanti ,  Abhishek Gupta

IPC: G06F3/06 ,  G06F9/455 ,  H04L29/06 ,  H04L29/08 ,  G06F16/174 ,  G06F16/23 ,  G06F16/25

Abstract: A logical group of data blocks stored in a first node is migrated to a second node according to a method that includes determining a first metric for each logical group of data blocks stored in the first node, the first metric representing a total size of the data blocks in the logical group, determining a second metric for each logical group of data blocks stored in the first node, the second metric representing a total size of the data blocks in the logical group that are uniquely stored in the first node, and selecting a logical group of data blocks for migration from the first node to the second node based on the first metric and the second metric.

公开(公告)号：US10268408B2

公开(公告)日：2019-04-23

申请号：US15853110

申请日：2017-12-22

Applicant: VMware, Inc.

Inventor： Mansi Shah ,  Eric Knauft ,  Jin Zhang ,  Christian Dickmann ,  Pascal Renauld ,  Radhika Vullikanti ,  Christos Karamanolis

IPC: G06F12/00 ,  G06F3/06 ,  G06F9/455

Abstract: In a storage cluster having nodes, blocks of a logical storage space of a storage object are allocated flexibly by a parent node to component nodes that are backed by physical storage. The method includes maintaining a first allocation map for the parent node, and second and third allocation maps for the first and second component nodes, respectively, executing a first write operation on the first component node and updating the second allocation map to indicate that the first block is a written block, selecting the second component node for executing a second write operation, and executing the second write operation on the second component node. Upon execution of the second write operation, the third allocation map is updated to indicate that the second block is a written block and the first allocation map is updated to indicate that the second block is allocated to the second component node.