公开(公告)号：US12067406B2

公开(公告)日：2024-08-20

申请号：US17819879

申请日：2022-08-15

Applicant: Xilinx, Inc.

Inventor： Baris Ozgul ,  David Clarke ,  Peter McColgan ,  Stephan Münz ,  Dylan Stuart ,  Pedro Miguel Parola Duarte ,  Juan J. Noguera Serra

IPC: G06F9/445 ,  G06F9/50 ,  G06F13/16 ,  G06F13/28 ,  G06F17/16 ,  G06N3/063

CPC classification number: G06F9/44505 ,  G06F9/5083 ,  G06F13/1673 ,  G06F13/28 ,  G06F17/16 ,  G06N3/063

Abstract: Using multiple overlays with a data processing array includes loading an application in a data processing array. The data processing array includes a plurality of compute tiles each having a processor. The application specifies kernels executable by the processors and implements stream channels that convey data to the plurality of compute tiles. During runtime of the application, a plurality of overlays are sequentially implemented in the data processing array. Each overlay implements a different mode of data movement in the data processing array via the stream channels. For each overlay implemented, a workload is performed by moving data to the plurality of compute tiles based on the respective mode of data movement.

公开(公告)号：US20230053537A1

公开(公告)日：2023-02-23

申请号：US17819879

申请日：2022-08-15

Applicant: Xilinx, Inc.

Inventor： Baris Ozgul ,  David Clarke ,  Peter McColgan ,  Stephan Munz ,  Dylan Stuart ,  Pedro Miguel Parola Duarte ,  Juan J. Noguera Serra

IPC: G06F13/16 ,  G06F13/28 ,  G06F17/16 ,  G06N3/063

Abstract: Using multiple overlays with a data processing array includes loading an application in a data processing array. The data processing array includes a plurality of compute tiles each having a processor. The application specifies kernels executable by the processors and implements stream channels that convey data to the plurality of compute tiles. During runtime of the application, a plurality of overlays are sequentially implemented in the data processing array. Each overlay implements a different mode of data movement in the data processing array via the stream channels. For each overlay implemented, a workload is performed by moving data to the plurality of compute tiles based on the respective mode of data movement.