Data processing method, apparatus, device, and readable storage medium

1. A data processing method applied to a solid-state drive, comprising:

acquiring historical input/output (I/O) data, wherein the historical I/O data comprises data of the solid-state drive accessed within a preset time period;

using a prediction model to learn the historical I/O data to obtain a prediction result, wherein the prediction result comprises an expected data access intensity of the solid-state drive in a future window period, and the future window period is determined according to a cycle in which the solid-state drive is accessed; and

managing the solid-state drive according to the prediction result, comprising:

responsive to the expected data access intensity of the solid-state drive being greater than a threshold, suspending a low-priority service request in the solid-state drive in the future window period; and

responsive to the expected data access intensity of the solid-state drive being smaller than the threshold, responding to the low-priority service request in the solid-state drive in the future window period,

wherein the data processing method further comprises:

responsive to the solid-state drive being accessed by a plurality of services, determining a corresponding cycle in which the solid-state drive is accessed by each service so that a plurality of cycles are determined for the plurality of services, respectively;

determining a maximum cycle among the plurality of cycles as a lower limit value of the future window period; and

determining a least common multiple of the plurality of cycles as an upper limit value of the future window period.

2. The data processing method according to claim 1 , further comprising:

responsive to the solid-state drive being accessed by only one service, determining a cycle in which the solid-state drive is accessed by the service as the future window period.

3. The data processing method according to claim 1 , wherein determining the future window period comprises:

determining a target value that is not less than the lower limit value and not greater than the upper limit value as the future window period.

4. The data processing method according to claim 1 , further comprising:

calculating a greatest common divisor of the plurality of cycles and determining the greatest common divisor as a prediction interval; and

determining a plurality of prediction points in the future window period according to the prediction interval.

5. The data processing method according to claim 1 , wherein using the prediction model to learn the historical I/O data to obtain the prediction result comprises:

using a long short-term memory (LSTM) model to learn the historical I/O data to obtain the prediction result.

6. The data processing method according to claim 1 , wherein managing the solid-state drive according to the prediction result comprises:

performing at least one of a cache management or garbage collection on the solid-state drive according to the prediction result.

7. A data processing device, comprising:

a memory configured to store a computer program; and

a processor, responsive to an execution of the computer program, configured to:

acquire historical input/output (I/O) data, wherein the historical I/O data comprises data of a solid-state drive accessed within a preset time period;

use a prediction model to learn the historical I/O data to obtain a prediction result, wherein the prediction result comprises an expected data access intensity of the solid-state drive in a future window period, and the future window period is determined according to a cycle in which the solid-state drive is accessed; and

manage the solid-state drive according to the prediction result by:

responsive to the expected data access intensity of the solid-state drive being greater than a threshold, suspending a low-priority service request in the solid-state drive in the future window period; and

responsive to the expected data access intensity of the solid-state drive being smaller than the threshold, responding to the low-priority service request in the solid-state drive in the future window period,

wherein the processor is further configured to:

responsive to the solid-state drive being accessed by a plurality of services, determine a corresponding cycle in which the solid-state drive is accessed by each service so that a plurality of cycles are determined for the plurality of services, respectively;

determine a maximum cycle among the plurality of cycles as a lower limit value of the future window period; and

determine a least common multiple of the plurality of cycles as an upper limit value of the future window period.

8. The data processing device according to claim 7 , wherein the processor is further configured to:

responsive to the solid-state drive being accessed by only one service, determine a cycle in which the solid-state drive is accessed by the service as the future window period.

9. The data processing device according to claim 7 , wherein to determine the future window period, the processor is further configured to:

determine a target value that is not less than the lower limit value and not greater than the upper limit value as the future window period.

10. The data processing device according to claim 7 , wherein the processor is further configured to:

calculate a greatest common divisor of the plurality of cycles and determine the greatest common divisor as a prediction interval; and

determine a plurality of prediction points in the future window period according to the prediction interval.

11. The data processing device according to claim 7 , wherein to use the prediction model to learn the historical I/O data to obtain the prediction result, the processor is configured to:

use a long short-term memory (LSTM) model to learn the historical I/O data to obtain the prediction result.

12. The data processing device according to claim 7 , wherein to manage the solid-state drive according to the prediction result, the processor is configured to:

perform at least one of a cache management or garbage collection on the solid-state drive according to the prediction result.

13. A non-transitory computer-readable storage medium, configured to store a computer program, wherein the computer program, when executed by a processor, causes the processor to perform operations, comprising:

acquiring historical input/output (I/O) data, wherein the historical I/O data comprises data of a solid-state drive accessed within a preset time period;

using a prediction model to learn the historical I/O data to obtain a prediction result, wherein the prediction result comprises an expected data access intensity of the solid-state drive in a future window period, and the future window period is determined according to a cycle in which the solid-state drive is accessed; and

managing the solid-state drive according to the prediction result, comprising:

responsive to the expected data access intensity of the solid-state drive being greater than a threshold, suspending a low-priority service request in the solid-state drive in the future window period; and

responsive to the expected data access intensity of the solid-state drive being smaller than the threshold, responding to the low-priority service request in the solid-state drive in the future window period,

wherein the operations further comprise:

responsive to the solid-state drive being accessed by a plurality of services, determining a corresponding cycle in which the solid-state drive is accessed by each service so that a plurality of cycles are determined for the plurality of services, respectively;

determining a maximum cycle among the plurality of cycles as a lower limit value of the future window period; and

determining a least common multiple of the plurality of cycles as an upper limit value of the future window period.

14. The computer-readable storage medium according to claim 13 , wherein the operations further comprise:

responsive to the solid-state drive being accessed by only one service, determining a cycle in which the solid-state drive is accessed by the service as the future window period.

15. The computer-readable storage medium according to claim 13 , wherein determining the future window period comprises:

determining a target value that is not less than the lower limit value and not greater than the upper limit value as the future window period.

16. The computer-readable storage medium according to claim 13 , wherein the operations further comprise:

calculating a greatest common divisor of the plurality of cycles and determining the greatest common divisor as a prediction interval; and

determining a plurality of prediction points in the future window period according to the prediction interval.

17. The computer-readable storage medium according to claim 13 , wherein using the prediction model to learn the historical I/O data to obtain the prediction result comprises:

using a long short-term memory (LSTM) model to learn the historical I/O data to obtain the prediction result.

18. The computer-readable storage medium according to claim 13 , wherein managing the solid-state drive according to the prediction result comprises:

performing at least one of a cache management or garbage collection on the solid-state drive according to the prediction result.