US 12,164,920 B2
Offloading data processing and knowledge synthesis
Viacheslav Dubeyko, Los Angeles, CA (US); and Jian Wang, Beijing (CN)
Assigned to Lemon Inc., Grand Cayman (KY); and Beijing Youzhuju Network Technology Co. Ltd., Beijing (CN)
Filed by Lemon Inc., Grand Cayman (KY); and Beijing Youzhuju Network Technology Co. Ltd., Beijing (CN)
Filed on Nov. 9, 2022, as Appl. No. 17/984,191.
Prior Publication US 2024/0152358 A1, May 9, 2024
Int. Cl. G06F 12/12 (2016.01); G06F 9/30 (2018.01); G06N 3/04 (2023.01)
CPC G06F 9/30047 (2013.01) [G06F 9/30032 (2013.01); G06F 9/30189 (2013.01); G06F 12/12 (2013.01); G06N 3/04 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method of offloading data processing and knowledge synthesis, comprising:
configuring a first memory for temporarily storing memory pages, wherein a set of flags are configured to indicate information about the memory pages, and wherein the set of flags are manageable by at least one central processing unit (CPU);
flushing a memory page among the memory pages from the first memory to a second memory based on determining that the memory page is associated with a first flag among the set of flags, wherein the second memory is configured to store a sequence of states of each of the memory pages, and wherein the first flag indicates that the memory page is ready to be flushed to the second memory;
determining data patterns and relations among the data patterns by data processing units (DPUs) based on the sequence of states of each of the memory pages in the second memory; and
building a knowledge base in a third memory based on the data patterns and the relations among the data patterns, wherein the third memory is a persistent memory.