公开(公告)号：US20220245453A1

公开(公告)日：2022-08-04

申请号：US17629437

申请日：2020-10-07

Applicant: Google LLC

Inventor： David Alexander Majnemer ,  Ravi Narayanaswami ,  Dong Hyuk Woo ,  Carrell Daniel Killebrew

IPC: G06N3/08

Abstract: Methods, systems, and apparatus, including an apparatus for redistributing tensor elements among computing units are described. In one aspect, a method includes distributing tensor elements of an N-dimensional tensor among multiple computing units of a computation system. Each computing unit redistributes the subset of tensor elements previously distributed to the computing unit to computing units. Each computing unit accesses redistribution partitioning data that specifies, for each computing unit, the tensor elements that are to be stored by the computing unit after redistributing the tensor elements. For each tensor element previously distributed to the particular computing unit, the computing unit determines a global linearized index value for the tensor element based on a multi-dimensional index for the tensor element. The computing unit determines, using the redistribution partitioning data and the global linearized index value, a destination computing unit and sends the tensor element to the destination computing unit.

公开(公告)号：US20210056396A1

公开(公告)日：2021-02-25

申请号：US16548555

申请日：2019-08-22

Applicant: Google LLC

Inventor： David Alexander Majnemer ,  Blake Alan Hechtman ,  Bjarke Hammersholt Roune

IPC: G06N3/063 ,  G06F17/16 ,  G06F17/15

Abstract: Methods and systems, including computer programs encoded on a computer storage medium. In one aspect, a method includes the actions of receiving a request to perform convolutional computations for a neural network on a hardware circuit having a matrix computation unit, the request specifying the convolutional computation to be performed on a feature tensor and a filter and padding applied to the feature tensor prior to performing the convolutional computation; and generating instructions that when executed by the hardware circuit cause the hardware circuit to perform operations comprising: transferring feature tensor data from a main memory of the hardware circuit to a scratchpad memory of the hardware circuit; and repeatedly performing the following operations: identifying a current subset of the feature tensor; and determining whether a memory view into the scratchpad memory for the current subset is consistent with a memory view of the current subset in the main memory.

公开(公告)号：US20200341807A1

公开(公告)日：2020-10-29

申请号：US16919968

申请日：2020-07-02

Applicant: Google LLC

Inventor： Yuanzhong Xu ,  James M. Stichnoth ,  David Alexander Majnemer

IPC: G06F9/48 ,  G06F9/38

Abstract: Methods, systems, and apparatus for scheduling first-in-first-out instructions are described. In one aspect, a method includes receiving data representing code of a program to be executed by a processing unit comprising hardware processors. For each of one or more of the hardware processors, an order of independent groups of first-in-first-out (FIFO) instructions for execution by the hardware processor is identified in the data representing the code of the program. For each independent group of FIFO instructions for execution by the hardware processor, a path length metric that represents how long it will take to reach an end of the program from the independent group of FIFO instructions is determined. A new order of the independent groups of FIFO instructions for execution by the hardware processor is generated based at least on the path length metric for each independent group of FIFO instructions for execution by the hardware processor.