公开(公告)号：US20200026601A1

公开(公告)日：2020-01-23

申请号：US16042409

申请日：2018-07-23

Applicant: Cisco Technology, Inc.

Inventor： Clint Ricker ,  Ryan Michael Okelberry ,  Dana Marlow Henriksen ,  Mel J. Oyler ,  Hoi-Tauw Jozef Chou ,  Kevin Wayne Kingdon

IPC: G06F11/10 ,  G06F11/07 ,  G06F3/06

Abstract: A disclosed method is performed at a fault-tolerant object-based storage system including M data storage entities, each is configured to store data on an object-basis. The method includes obtaining a request to store N copies of a data object and in response, storing the N copies of the data object across the M data storage entities, where the N copies are distributed across the M data storage entities. The method additionally includes generating a first parity object for a first subset of M copies of the N copies of the data object, where the first parity object is stored on a first parity storage entity separate from the M data storage entities. The method also includes generating a manifest linking the first parity object with one or more other subsets of M copies of the N copies of the data object.

公开(公告)号：US20190265895A1

公开(公告)日：2019-08-29

申请号：US16017697

申请日：2018-06-25

Applicant: Cisco Technology, Inc.

Inventor： Ryan Michael Okelberry ,  Dana Henriksen ,  Mel J. Oyler ,  Hoi-Tauw Jozef Chou ,  Kevin Wayne Kingdon

IPC: G06F3/06 ,  G06F11/30 ,  G11B20/12

Abstract: Various implementations of hard disk track management method, device, and system disclosed herein enable improvements of file system write bandwidth. In various implementations, a method is performed at a disk storage including a file controller controlling a disk drive with a disk platter that is divided into multiple regions including a fast region. In various implementations, the method includes receiving a write request associated with data to be written to the disk drive and in response, determining a disk utilization of the disk drive. In various implementations, the method further includes placing the disk drive in a surge mode to write the data to the fast region upon determining that the disk utilization is above a first threshold, and placing the disk drive in a non-surge mode to write the data to other regions of the multiple regions upon determining that the disk utilization is below a second threshold.

公开(公告)号：US10782888B2

公开(公告)日：2020-09-22

申请号：US16017697

申请日：2018-06-25

Applicant: Cisco Technology, Inc.

Inventor： Ryan Michael Okelberry ,  Dana Henriksen ,  Mel J. Oyler ,  Hoi-Tauw Jozef Chou ,  Kevin Wayne Kingdon

IPC: G06F3/06 ,  G11B20/12 ,  G06F11/30

Abstract: Various implementations of hard disk track management method, device, and system disclosed herein enable improvements of file system write bandwidth. In various implementations, a method is performed at a disk storage including a file controller controlling a disk drive with a disk platter that is divided into multiple regions including a fast region. In various implementations, the method includes receiving a write request associated with data to be written to the disk drive and in response, determining a disk utilization of the disk drive. In various implementations, the method further includes placing the disk drive in a surge mode to write the data to the fast region upon determining that the disk utilization is above a first threshold, and placing the disk drive in a non-surge mode to write the data to other regions of the multiple regions upon determining that the disk utilization is below a second threshold.

公开(公告)号：US10802914B2

公开(公告)日：2020-10-13

申请号：US16042409

申请日：2018-07-23

Applicant: Cisco Technology, Inc.

Inventor： Clint Ricker ,  Ryan Michael Okelberry ,  Dana Marlow Henriksen ,  Mel J. Oyler ,  Hoi-Tauw Jozef Chou ,  Kevin Wayne Kingdon

IPC: G06F11/00 ,  G06F11/10 ,  G06F11/07 ,  G06F3/06

Abstract: A disclosed method is performed at a fault-tolerant object-based storage system including M data storage entities, each is configured to store data on an object-basis. The method includes obtaining a request to store N copies of a data object and in response, storing the N copies of the data object across the M data storage entities, where the N copies are distributed across the M data storage entities. The method additionally includes generating a first parity object for a first subset of M copies of the N copies of the data object, where the first parity object is stored on a first parity storage entity separate from the M data storage entities. The method also includes generating a manifest linking the first parity object with one or more other subsets of M copies of the N copies of the data object.