US 12,340,300 B1
Streaming processor architecture
Dennis Charles Abts, Mountain View, CA (US); and Jonathan Alexander Ross, Mountain View, CA (US)
Assigned to Groq, Inc., Mountain View, CA (US)
Filed by Groq, Inc., Mountain View, CA (US)
Filed on Mar. 16, 2021, as Appl. No. 17/203,214.
Application 17/203,214 is a continuation of application No. 17/103,910, filed on Nov. 24, 2020, abandoned.
Application 17/203,214 is a continuation of application No. 16/928,958, filed on Jul. 14, 2020, granted, now 11,307,827, issued on Apr. 19, 2022.
Application 17/203,214 is a continuation of application No. 16/132,196, filed on Sep. 14, 2018, granted, now 11,114,138.
Application 17/203,214 is a continuation of application No. 16/932,632, filed on Jul. 17, 2020, granted, now 11,165,428.
Application 17/203,214 is a continuation of application No. 16/243,768, filed on Jan. 9, 2019, granted, now 11,115,147.
Application 17/203,214 is a continuation of application No. 16/951,938, filed on Nov. 18, 2020.
Application 17/203,214 is a continuation of application No. 16/526,916, filed on Jul. 30, 2019, granted, now 11,210,594.
Application 17/203,214 is a continuation of application No. 16/561,516, filed on Sep. 5, 2019, granted, now 11,136,116.
Application 17/203,214 is a continuation of application No. 16/526,936, filed on Jul. 30, 2019, granted, now 11,568,275.
Application 17/203,214 is a continuation of application No. 16/277,817, filed on Feb. 15, 2019, granted, now 10,908,900.
Application 17/203,214 is a continuation of application No. 16/686,858, filed on Nov. 18, 2019, granted, now 11,455,370.
Application 17/203,214 is a continuation of application No. 16/686,864, filed on Nov. 18, 2019, granted, now 11,204,976.
Application 17/203,214 is a continuation of application No. 16/686,866, filed on Nov. 18, 2019, granted, now 11,301,546.
Application 17/203,214 is a continuation of application No. 16/686,870, filed on Nov. 18, 2019, granted, now 11,537,687.
Int. Cl. G06N 3/063 (2023.01); G06F 9/30 (2018.01); G06F 18/214 (2023.01); G06N 3/08 (2023.01)
CPC G06N 3/063 (2013.01) [G06F 9/3001 (2013.01); G06F 9/30036 (2013.01); G06F 18/214 (2023.01); G06N 3/08 (2013.01)] 10 Claims
OG exemplary drawing
 
1. A Tensor Streaming Processor (TSP) that transforms streams of data representing arrays of numerical information with one or more dimensions, comprising:
at least one host-to-processor communication module for transferring the data between a host computer and the processor; and
a plurality of superlanes, each superlane of the plurality of superlanes being partitioned into slices, wherein each slice of the slices comprises a set of tiles with the same functionality executing the same instructions, located in similar positions in different superlanes of the plurality of superlanes, wherein the instructions are supplied from buffers that comprise an Instruction Control Unit (ICU), and wherein a set of data processed by one superlane of the plurality of superlanes comprises a stream of data,
wherein each superlane of the plurality of superlanes comprises a plurality of functional tiles, wherein data output of a tile in a first set of functional tiles is bidirectionally transferred to data input of a tile in a second set of functional tiles, and wherein the data output of a tile in the second set of functional tiles is bidirectionally transferred to the data input of a tile in the first set of functional tiles.