Data storage device and method for multi-level conditional prediction of future random read commands
A data storage device and method for multi-level conditional prediction of future random read commands are provided. In one embodiment, a data storage device is provided comprising a memory and a controller. The controller is configured to receive a random read command from a host, wherein the received random read command is associated with a stream; predict a next stream to be received from the host; and predict a next random read command to be received from the host based on the received random read command and the predicted next stream. Other embodiments are possible, and each of the embodiments can be used alone or together in combination.
1. A data storage device comprising:
a memory; and
a controller configured to:
receive a random read command from a host, wherein the received random read command is associated with a stream;
predict a next stream to be received from the host;
predict a next random read command to be received from the host based on the received random read command and the predicted next stream; and
pre-fetch, from the memory, data associated with the predicted next random read command.
2. The data storage device of claim 1 , wherein the stream and the predicted next stream are associated with different applications.
3. The data storage device of claim 1 , wherein the controller is further configured to predict the next stream using a stream prediction model.
4. The data storage device of claim 3 , wherein the stream prediction model comprises a machine-learning-based prediction model.
5. The data storage device of claim 4 , wherein the machine-learning-based prediction model is configured to receive workload statistics as an input.
6. The data storage device of claim 4 , wherein the machine-learning-based prediction model is configured to receive power consumption as an input.
7. The data storage device of claim 3 , wherein the stream prediction model comprises a statistical-based prediction model.
8. The data storage device of claim 7 , wherein the statistical-based prediction model is configured to use counters to track streams.
9. The data storage device of claim 1 , wherein the controller is further configured to use a history-pattern-matcher (HPM) to predict the next random read command.
10. The data storage device of claim 1 , wherein the memory comprises a three-dimensional memory.
11. The data storage device of claim 1 , wherein the stream and the next stream are associated with different ones of at least some of the following: namespaces or zones submission queues.
12. The data storage device of claim 1 , wherein the stream and the next stream are associated with different ones of at least some of the following: user identifiers, host identifiers, or stream identifiers.
13. A method comprising:
performing the following in a data storage device in communication with a host:
receiving a random read command from the host;
using a multi-level random read prediction method to predict a future random read command from the host and
pre-fetching, from a memory in the data storage device, data associated with the predicted future random read command;
wherein:
a first level of the multi-level random read prediction method identifies a stream associated with the received random read command and predicts a future stream from the host; and
a second level of the multi-level random read prediction method predicts the future random read command using the received random read command and the predicted future stream.
14. The method of claim 13 , wherein the first level of the multi-level random read prediction method comprises a training/learning stage that tracks and learns repeating patterns of streams.
15. The method of claim 13 , wherein the second level of the multi-level random read prediction method comprises an inference/prediction stage that learns repeating random read patterns.
16. The method of claim 13 , wherein the stream and the predicted future stream are associated with different ones of at least some of the following: applications, namespaces, user identifiers, host identifiers, zones submission queues, or stream identifiers.
17. The method of claim 13 , wherein the multi-level random read prediction method comprises a machine-learning model.
18. The method of claim 13 , wherein the multi-level random read prediction method comprises a statistical-based model.
19. The method of claim 13 , wherein the multi-level random read prediction method comprises a history-pattern-matcher (HPM).
20. A data storage device comprising:
a memory;
means for receiving a random read command from a host, wherein the received random read command is associated with a stream;
means for predicting a next stream to be received from the host;
means for predicting a next random read command to be received from the host based on the received random read command and the predicted next stream; and
means for pre-fetching, from the memory, data associated with the predicted next random read command.