公开(公告)号：US20200326970A1

公开(公告)日：2020-10-15

申请号：US16384769

申请日：2019-04-15

Applicant: Microsoft Technology Licensing, LLC

Inventor： Bryan Stephen MATTHEW ,  Scott Chao-Chueh LEE

IPC: G06F9/455 ,  G06F3/06 ,  G06F16/18

Abstract: A request is received at an interface configured to receive storage requests for storing data in the software-defined storage network using an append-only storage scheme. The requests include an identifier of a data object to be stored in the software-defined storage network. The requests are agnostic of interfaces and hardware-specific details of the storage devices of the software-defined storage network. A virtualization layer of the software-defined storage network accesses data indicative of space allocation for the plurality of storage devices; and policies for prioritizing performance of the software-defined storage network. Based on the data and policies, a physical storage location at the plurality of storage devices is selected for storing the data object. Metadata is generated for the data object indicating that the data object is an append-only object and mapping the physical storage location of the data object to the identifier. The request is translated to one or more instructions for storing the data object at the physical storage location using the append-only storage scheme. The data object is caused to be stored at the physical storage location using the append-only storage scheme. Data usable to render status information indicating that the data object has been appended is returned.

公开(公告)号：US20230089663A1

公开(公告)日：2023-03-23

申请号：US17800517

申请日：2021-03-15

Applicant: Microsoft Technology Licensing, LLC

Inventor： Vinod R. SHANKAR ,  Scott Chao-Chueh LEE ,  Bryan Stephen MATTHEW

IPC: G06F3/06

Abstract: A reduced throughput maintenance mode for adaptively managing input/output (I/O) operations within a resilient group of storage nodes. A first storage node in a resilient group of storage nodes is classified as operating in a normal throughput mode, and a second storage node in the resilient group is classified as operating in a reduced throughput mode. While the second node is classified as operating in the reduced throughput mode, read and write I/O operations are queued for the resilient group. The read I/O operation is prioritized for assignment to the first node, so as to reduce I/O load on the second node while it operates in the reduced throughput mode. The write I/O operation is queued to the second node, so as to maintain synchronization of the second node with the resilient group while it operates in the reduced throughput mode.

公开(公告)号：US20200349121A1

公开(公告)日：2020-11-05

申请号：US16399910

申请日：2019-04-30

Applicant: Microsoft Technology Licensing, LLC

Inventor： Scott Chao-Chueh LEE ,  Rajsekhar DAS ,  Bryan Stephen MATTHEW

IPC: G06F16/18

Abstract: A file system is implemented in a system comprising an anonymous-write storage device. A request for a file system operation that comprises semantics of the file system to receiving the request is received. The file system translates the file system operation to one or more commands that effect the file system operation while maintaining consistency with operation of the anonymous-write storage device. The one or more commands comprise one or more of updating file system metadata, sending a request to the anonymous-write storage device, or receiving a response from the anonymous-write storage device. The anonymous-write storage device is configured to write data in an append-only format and determine locations of data write operations agnostic of input of a write location from the file system. Metadata is updated to indicate results of the file system operation.

公开(公告)号：US20190303243A1

公开(公告)日：2019-10-03

申请号：US15943555

申请日：2018-04-02

Applicant: Microsoft Technology Licensing, LLC

Inventor： Scott Chao-Chueh LEE ,  Aaron William OGUS ,  Subhasish BHATTACHARYA ,  Bryan Stephen MATTHEW ,  Srinivasan MALAYALA

IPC: G06F11/10 ,  G06F3/06 ,  G06F11/20 ,  G06F11/16

Abstract: The techniques described herein are configured to improve the remanufacturing process by waiting to rebuild selective data items stored on a storage device that is unavailable. A storage device is unavailable when it is taken offline and/or disconnected from a network. The storage device may be taken offline due to a failed component (e.g., an actuator arm, an actuator head, damage to the underlying storage media, etc.). The storage device comprises multiple independent logical units, where a logical unit is a uniquely addressable portion of a storage device that is visible and/or that is exposed to a host. Accordingly, the techniques described herein are configured to conserve resources by selectively rebuilding data items stored in logical units of an offline storage device that are not affected by a remanufacturing process implemented due to a failed component.