The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.
The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.
Patent No.:
Date of Patent:
Jan. 21, 2025
Filed:
May. 25, 2023
Applicant:
Sambanova Systems, Inc., Palo Alto, CA (US);
Inventor:
Maulik Desai, Cedar Park, TX (US);
Assignee:
SambaNova Systems, Inc., Palo Alto, CA (US);
Attorneys:
Primary Examiner:
Int. Cl.
CPC ...
G06F 15/76 (2006.01); G06F 15/78 (2006.01); G06F 17/18 (2006.01); G06N 3/02 (2006.01); G06N 3/08 (2023.01);
U.S. Cl.
CPC ...
G06F 15/7867 (2013.01); G06F 17/18 (2013.01); G06N 3/02 (2013.01); G06N 3/08 (2013.01);
Abstract
The softmax operation is pipelined to evenly balanced operations with 2N latency per sharded M dimension of a tensor shaped M*N, resulting in ˜1.8× performance gain. As each operation is fragmented, the pipeline does not assume a fixed-cost fill that can performance-wise hurt significantly for small Tensor dimensions. In addition, this innovative design is Place-and-Route (PNR) friendly as well as resource efficient. Moreover, it is easily parallelized without requiring additional support at the subnet level.