公开(公告)号：US20220012305A1

公开(公告)日：2022-01-13

申请号：US17485055

申请日：2021-09-24

Applicant: Intel Corporation

Inventor： Dan BAUM ,  Chen KOREN ,  Elmoustapha OULD-AHMED-VALL ,  Michael ESPIG ,  Christopher J. HUGHES ,  Raanan SADE ,  Robert VALENTINE ,  Mark J. CHARNEY ,  Alexander F. HEINECKE

IPC: G06F17/16 ,  G06F9/38 ,  G06F9/30

Abstract: Disclosed embodiments relate to accelerating multiplication of sparse matrices. In one example, a processor is to fetch and decode an instruction having fields to specify locations of first, second, and third matrices, and an opcode indicating the processor is to multiply and accumulate matching non-zero (NZ) elements of the first and second matrices with corresponding elements of the third matrix, and executing the decoded instruction as per the opcode to generate NZ bitmasks for the first and second matrices, broadcast up to two NZ elements at a time from each row of the first matrix and each column of the second matrix to a processing engine (PE) grid, each PE to multiply and accumulate matching NZ elements of the first and second matrices with corresponding elements of the third matrix. Each PE further to store an NZ element for use in a subsequent multiplications.

公开(公告)号：US20200210517A1

公开(公告)日：2020-07-02

申请号：US16234374

申请日：2018-12-27

Applicant: Intel Corporation

Inventor： Dan BAUM ,  Chen KOREN ,  Elmoustapha OULD-AHMED-VALL ,  Michael ESPIG ,  Christopher J. HUGHES ,  Raanan SADE ,  Robert VALENTINE ,  Mark J. CHARNEY ,  Alexander F. HEINECKE

IPC: G06F17/16 ,  G06F9/38 ,  G06F9/30

Abstract: Disclosed embodiments relate to accelerating multiplication of sparse matrices. In one example, a processor is to fetch and decode an instruction having fields to specify locations of first, second, and third matrices, and an opcode indicating the processor is to multiply and accumulate matching non-zero (NZ) elements of the first and second matrices with corresponding elements of the third matrix, and executing the decoded instruction as per the opcode to generate NZ bitmasks for the first and second matrices, broadcast up to two NZ elements at a time from each row of the first matrix and each column of the second matrix to a processing engine (PE) grid, each PE to multiply and accumulate matching NZ elements of the first and second matrices with corresponding elements of the third matrix. Each PE further to store an NZ element for use in a subsequent multiplications.

公开(公告)号：US20240078285A1

公开(公告)日：2024-03-07

申请号：US18502291

申请日：2023-11-06

Applicant: Intel Corporation

Inventor： Dan BAUM ,  Chen KOREN ,  Elmoustapha OULD-AHMED-VALL ,  Michael ESPIG ,  Christopher J. HUGHES ,  Raanan SADE ,  Robert VALENTINE ,  Mark J. CHARNEY ,  Alexander F. HEINECKE

IPC: G06F17/16 ,  G06F9/30 ,  G06F9/38

CPC classification number: G06F17/16 ,  G06F9/3001 ,  G06F9/30101 ,  G06F9/3016 ,  G06F9/3802

Abstract: Disclosed embodiments relate to accelerating multiplication of sparse matrices. In one example, a processor is to fetch and decode an instruction having fields to specify locations of first, second, and third matrices, and an opcode indicating the processor is to multiply and accumulate matching non-zero (NZ) elements of the first and second matrices with corresponding elements of the third matrix, and executing the decoded instruction as per the opcode to generate NZ bitmasks for the first and second matrices, broadcast up to two NZ elements at a time from each row of the first matrix and each column of the second matrix to a processing engine (PE) grid, each PE to multiply and accumulate matching NZ elements of the first and second matrices with corresponding elements of the third matrix. Each PE further to store an NZ element for use in a subsequent multiplications.

公开(公告)号：US20200258890A1

公开(公告)日：2020-08-13

申请号：US16859600

申请日：2020-04-27

Applicant: Intel Corporation

Inventor： Charles AUGUSTINE ,  Somnath PAUL ,  Muhammad M. KHELLAH ,  Chen KOREN

IPC: H01L27/11 ,  G11C11/412 ,  G11C11/419 ,  G11C11/418

Abstract: An ultra-deep compute Static Random Access Memory (SRAM) with high compute throughput and multi-directional data transfer capability is provided. Compute units are placed in both horizontal and vertical directions to achieve a symmetric layout while enabling communication between the compute units. An SRAM array supports simultaneous read and write to the left and right section of the same SRAM subarray by duplicating pre-decoding logic inside the SRAM array. This allows applications with non-overlapping read and write address spaces to have twice the bandwidth as compared to a baseline SRAM array.