公开(公告)号：US10402111B1

公开(公告)日：2019-09-03

申请号：US15676083

申请日：2017-08-14

Applicant: Xilinx, Inc.

Inventor： Deboleena Sakalley ,  Ramesh R. Subramanian ,  Gopikrishna Jandhyala ,  Santosh Singh ,  Seong Hwan Kim

IPC: G06F3/06 ,  G06F13/42

Abstract: A data storage system includes a bridging device. The bridging device is configured to receive, from a host through a network, a host data block size. A sub-block size is determined based on the host data block size. One or more storage devices are configured to include a plurality of storage sub-blocks each having the sub-block size. A first write command to write first host data including a first number of host data blocks to the one or more storage devices is received. The bridging device compresses the first host data to generate first compressed data, and write the first compressed data to a second number of storage sub-blocks of the one or more storage devices.

公开(公告)号：US09934173B1

公开(公告)日：2018-04-03

申请号：US15052603

申请日：2016-02-24

Applicant: Xilinx, Inc.

Inventor： Deboleena Sakalley ,  Santosh Singh ,  Ramesh R. Subramanian ,  Pankaj V. Kumbhare ,  Ravi K. Boddu

IPC: G06F13/16 ,  G06F13/28 ,  G06F13/40

CPC classification number: G06F13/1673 ,  G06F13/28 ,  G06F13/4068

Abstract: An example method of exchanging data between a remote host and a target system includes receiving at least one remote descriptor from the remote host over a front-end fabric at a controller, the at least one remote descriptor specifying a remote buffer in a remote memory of the remote host that is larger than a page size. The method includes adding entries to a table that map the remote buffer to a plurality of page-sized virtual buffers in a virtual address space managed by the controller, generating local descriptors referencing the plurality of paged-sized virtual buffers, receiving a sequence of page-sized direct memory access (DMA) requests at the controller, generating a sequence remote DMA (RDMA) requests from the sequence of DMA requests based on the entries in the table, and sending the sequence of RDMA requests to the remote host over the front-end fabric.

公开(公告)号：US11861326B1

公开(公告)日：2024-01-02

申请号：US15092490

申请日：2016-04-06

Applicant: Xilinx, Inc.

Inventor： Santosh Singh ,  Deboleena M. Sakalley ,  Ramesh R. Subramanian ,  Pankaj V. Kumbhare ,  Ravi K. Boddu

IPC: G06F15/16 ,  G06F5/06 ,  G06F3/06 ,  G06F15/173

CPC classification number: G06F5/065 ,  G06F3/061 ,  G06F3/0655 ,  G06F3/0688 ,  G06F15/17331 ,  G06F2205/067

Abstract: An example method of flow control between remote hosts and a target system over a front-end fabric, the target system including a nonvolatile memory (NVM) subsystem coupled to a back end fabric having a different transport than the front-end fabric is described. The method includes receiving commands from the remote hosts at a controller in the target system for the NVM subsystem. The method further includes storing the commands in a first-in-first-out (FIFO) shared among the remote hosts and implemented in memory of the target system. The method further includes updating virtual submission queues for the remote hosts based on the commands stored in the FIFO. The method further includes providing the commands to the NVM subsystem from the FIFO.