Memory device with microbumps to transmit data for a machine learning operation

1. A system comprising:

a non-volatile memory device operatively coupled to a processing device, the non-volatile memory device to store data for a machine learning operation, the non-volatile memory device comprising a first plurality of interconnect components, wherein:

the processing device is to select, based at least in part on a condition of the machine learning operation, a set of interconnect components of the first plurality of interconnect components to transmit the data for the machine learning operation from the non-volatile memory device to a second plurality of interconnect components of a machine learning processing device to perform the machine learning operation.

2. The system of claim 1 , wherein a respective interconnect component of the first plurality of interconnect components and a respective interconnect component of the second plurality of interconnect components are coupled via an interface layer.

3. The system of claim 2 , wherein the non-volatile memory device is further coupled to a volatile memory device via the interface layer and a plurality of input and output (I/O) pins of the volatile memory device.

4. The system of claim 1 , wherein the data stored in the non-volatile memory device is model data and weight data, the model data defining the machine learning operation and the weight data representing one or more numerical value used in the machine learning operation.

5. The system of claim 1 , wherein a respective interconnect component of the first plurality of interconnect components and a respective interconnect component of the second plurality of interconnect components are coupled via an interface layer.

6. The system of claim 5 , wherein the volatile memory device is further coupled to a non-volatile memory device via the interface layer and a plurality of input and output (I/O) pins of the non-volatile memory device.

7. A system comprising:

a volatile memory device operatively coupled to a processing device, the volatile memory device to store data for a machine learning operation, the volatile memory device comprising a first plurality of interconnect components, wherein:

the processing device is to select, based at least in part on a condition of the machine learning operation, a set of interconnect components of the first plurality of interconnect components to transmit the data for the machine learning operation from the volatile memory device to a second plurality of interconnect components of a machine learning processing device to perform the machine learning operation.

8. The system of claim 7 , wherein the data stored in the volatile memory device includes input data, the input data corresponding to data to be processed by the machine learning operation.

9. A method comprising:

receiving, by a processing device, a command to transmit data for a machine learning operation;

selecting, by the processing device based at least in part on a condition of the machine learning operation, a set of interconnect components of a first memory device to be used to transmit the data to a machine learning processing device to perform the machine learning operation; and

transmitting, by the processing device, the data from the first memory device to the machine learning processing device via the set of interconnect components.

10. The method of claim 9 , wherein the determining of selecting the set of interconnect components comprises:

determining that the data is stored on the first memory device; and

responsive to determining that the data is stored on the first memory device, determining the set of interconnect components of the first memory device.

11. The method of claim 9 , wherein selecting the set of interconnect components further comprises:

determining that the data is stored on a second memory device;

responsive to determining that the data is stored on the second memory device, retrieving the data from the second memory device, the second memory device having a plurality of input and output (I/O) pins coupled with the first memory device; and

storing the data to the first memory device.

12. The method of claim 11 , wherein the first memory device corresponds to a volatile memory device and the second memory device corresponds to a non-volatile memory device.

13. The method of claim 11 , wherein the data corresponds to at least one of model data and weight data for the machine learning operation, the model data defining the machine learning operation and the weight data representing one or more numerical value used in the machine learning operation.

14. The method of claim 9 , wherein selecting the set of interconnect components further comprises:

determining that the data is stored on a second memory device; and

responsive to determining that the data is stored on the second memory device, causing the second memory device to transmit the data to the machine learning processing device, the second memory device having a plurality of I/O pins coupled with the machine learning processing device.

15. The method of claim 14 , wherein the first memory device corresponds to a non-volatile memory device and the second memory device corresponds to a volatile memory device.

16. The method of claim 9 , wherein the data corresponds to input data for the machine learning operation, the input data corresponding to data to be processed by the machine learning operation.

17. The method of claim 9 , wherein the transmitting of the data comprises transmitting the data from the set of interconnect components of the memory device to corresponding set of interconnect components of the machine learning processing device.

18. The method of claim 17 , wherein a respective interconnect component in the set of interconnect components of the first memory device is coupled with a respective interconnect component of the corresponding set of interconnect components of the machine learning processing device.