Transmission of data for a machine learning operation using different microbumps
A system includes a memory device with microbumps and a processing device. The processing device is operatively coupled with the memory device to perform operations. The operations include receiving information indicating a current condition of the machine learning operation while data for the machine learning operation is being transmitted using a first set of microbumps of the plurality of microbumps. Furthermore, the operations include, in response to a change in the condition of the machine learning operation, transmitting subsequent data using a second set of microbumps of the plurality of microbumps, wherein a number of microbumps included in the second set of microbumps is based on the received information.
1. A system comprising:
a memory device comprising a plurality of microbumps; and
a processing device, operatively coupled with the memory device, to perform operations comprising:
receiving information indicating a current condition of a machine learning operation while data for the machine learning operation is being transmitted using a first set of microbumps of the plurality of microbumps; and
in response to a change in the condition of the machine learning operation, transmitting subsequent data using a second set of microbumps of the plurality of microbumps, wherein a number of microbumps included in the second set of microbumps is based on the received information.
2. The system of claim 1 , wherein the condition of the machine learning operation includes at least one of a status of a power supply associated with the memory device, a temperature associated with the memory device, or a data size associated with the machine learning operation.
3. The system of claim 1 , wherein the number of microbumps included the second set of microbumps is based on the change in the condition of the machine learning operation.
4. The system of claim 3 , wherein the number of microbumps included in the second set of microbumps is based on a direction of the change in the condition of the machine learning operation.
5. The system of claim 3 , wherein the number of microbumps included in the second set of microbumps is based on a magnitude of the change in the condition of the machine learning operation.
6. The system of claim 1 , wherein the second set of microbumps includes groups of microbumps in different locations than the locations of the groups of microbumps included in the first set of microbumps.
7. The system of claim 1 , wherein the memory device corresponds to a non-volatile memory device.
8. The system of claim 1 , wherein the memory device corresponds to a volatile memory device.
9. A method comprising:
receiving information indicating a current condition of a machine learning operation while data for the machine learning operation is being transmitted using a first set of microbumps of a plurality of microbumps on a memory device; and
in response to a change in the condition of the machine learning operation, transmitting subsequent data using a second set of microbumps of the plurality of microbumps, wherein a number of microbumps included in the second set of microbumps is based on the received information.
10. The method of claim 9 , wherein the condition of the machine learning operation includes at least one of a status of a power supply associated with the memory device, a temperature associated with the memory device, or a data size associated with the machine learning operation.
11. The method of claim 9 , wherein the number of microbumps included the second set of microbumps is on the change in the condition of the machine learning operation.
12. The method of claim 11 , wherein the number of microbumps included in the second set of microbumps is based on a direction of the change in the condition of the machine learning operation.
13. The method of claim 11 , wherein the number of microbumps included in the second set of microbumps is based on a magnitude of the change in the condition of the machine learning operation.
14. The method of claim 11 , wherein the second set of microbumps includes groups of microbumps in different locations than the locations of the groups of microbumps included in the first set of microbumps.
15. The method of claim 9 , wherein the memory device corresponds to a non-volatile memory device.
16. The method of claim 9 , wherein the memory device corresponds to a volatile memory device.
17. A non-transitory computer readable storage medium comprising instructions that, when executed by a processing device operatively coupled with a memory device, cause the processing device to perform operations comprising:
receiving an indication to change a first set of microbumps in a memory device to transmit data for a machine learning operation, the indication comprising information about a condition of a machine learning operation;
determining a total number of microbumps to be selected based on the indication;
selecting, by a processing device, a second set of microbumps based on locations of respective microbumps in the memory device in accordance with the indication; and
transmitting the data for the machine learning operation using the second set of microbumps.
18. The non-transitory computer readable storage medium of claim 17 , wherein the selecting of the second set of the microbumps comprises:
determining a plurality of groups of microbumps, each group comprising a defined number of microbumps and each group being adjacent to another group of the plurality of groups of microbumps in the memory device; and
identifying the second set of microbumps corresponding to the total number of microbumps from the plurality of groups of microbumps in the memory device.
19. The non-transitory computer readable storage medium of claim 18 , wherein the identifying of the second set of the microbumps comprises:
determining a second number of microbumps for each group based on the total number of microbumps and a total number of groups in the plurality of groups of microbumps; and
for each group in the plurality of groups of microbumps, selecting a subset of microbumps comprising the second number of microbumps.
20. The non-transitory computer readable storage medium of claim 18 , wherein the identifying of the second set of the microbumps comprises:
determining a second number of groups for selection based on the total number of microbumps to be selected and a number of microbumps in each group;
selecting a subset of groups corresponding to the second number of groups; and
determining the second set of microbumps including each microbump in the subset of groups.