US 11,704,599 B2
System for performing a machine learning operation using microbumps
Poorna Kale, Folsom, CA (US)
Assigned to Micron Technology, Inc., Boise, ID (US)
Filed by Micron Technology, Inc., Boise, ID (US)
Filed on Dec. 4, 2019, as Appl. No. 16/703,122.
Prior Publication US 2021/0174260 A1, Jun. 10, 2021
Int. Cl. G06N 20/10 (2019.01); G06F 3/06 (2006.01); G06N 20/00 (2019.01)
CPC G06N 20/10 (2019.01) [G06F 3/0604 (2013.01); G06F 3/0658 (2013.01); G06F 3/0685 (2013.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
a machine learning processing device to perform a machine learning operation, the machine learning processing device comprising a first plurality of microbumps;
a memory device to store data for the machine learning operation, the memory device comprising a second plurality of microbumps, wherein:
the first plurality of microbumps of the machine learning processing device are coupled with the second plurality of microbumps of the memory device; and
the first plurality of microbumps of the machine learning processing device and the second plurality of microbumps of the memory device are to transmit the data for the machine learning operation; and
a processing device to identify a subset of the second plurality of microbumps of the memory device for data transfer based on a received indication, wherein the indication specifies a change regarding the subset of microbumps.
 
10. A system comprising:
a machine learning processing device to perform a machine learning operation;
a volatile memory device to store first data for the machine learning operation, the volatile memory device comprising a first plurality of microbumps that transmit the first data to the machine learning processing device;
a non-volatile memory device to store second data for the machine learning operation, the non-volatile memory device comprising a second plurality of microbumps that transmit the second data to the machine learning processing device; and
a processing device to determine a set of microbumps of a respective memory device corresponding to one of the non-volatile memory device or the volatile memory device.
 
15. A method comprising:
receiving a command to transmit data for a machine learning operation to a machine learning processing device, wherein the machine learning processing device performs the machine learning operation using the data;
determining, by a processing device, whether the data is stored in a first memory device or a second memory device;
determining a set of microbumps of a respective memory device corresponding to one of the first memory device or the second memory device storing the data; and
transmitting the data to the machine learning processing device using the set of microbumps of the respective memory device.