US 12,380,060 B2
Graph spatial split
Yun Du, Palo Alto, CA (US); Gao Deng, Palo Alto, CA (US); Jianding Luo, Palo Alto, CA (US); and Zhengyu Chen, Palo Alto, CA (US)
Assigned to SambaNova Systems, Inc., Palo Alto, CA (US)
Filed by SambaNova Systems, Inc., Palo Alto, CA (US)
Filed on May 25, 2023, as Appl. No. 18/202,059.
Claims priority of provisional application 63/348,961, filed on Jun. 3, 2022.
Claims priority of provisional application 63/346,234, filed on May 26, 2022.
Claims priority of provisional application 63/345,740, filed on May 25, 2022.
Prior Publication US 2024/0168915 A1, May 23, 2024
Int. Cl. G06F 15/82 (2006.01); G06F 9/38 (2018.01)
CPC G06F 15/825 (2013.01) [G06F 9/3867 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A system for reducing latency and increasing throughput in reconfigurable dataflow processors, the system comprising:
a host computer comprising a graph optimization module configured to conduct a method comprising:
receiving a compute graph for execution on a reconfigurable dataflow computing system, the compute graph comprising a node specifying an operation on a tensor;
splitting the node into multiple nodes that each specify the operation on a distinctive portion of the tensor to produce a first modified compute graph; and
a reconfigurable dataflow processor (RDP) configured to execute the first modified compute graph.