US 12,079,476 B2
Data processing method, apparatus, device, and readable storage medium
Yan Wang, Shenzhen (CN); and Weijun Li, Shenzhen (CN)
Assigned to SHENZHEN DAPU MICROELECTRONICS CO., LTD., Shenzhen (CN)
Filed by SHENZHEN DAPU MICROELECTRONICS CO., LTD., Shenzhen (CN)
Filed on Apr. 29, 2022, as Appl. No. 17/733,225.
Application 17/733,225 is a continuation of application No. PCT/CN2020/102020, filed on Jul. 15, 2020.
Claims priority of application No. 201911077491.9 (CN), filed on Nov. 6, 2019.
Prior Publication US 2022/0253214 A1, Aug. 11, 2022
Int. Cl. G06F 3/06 (2006.01)
CPC G06F 3/061 (2013.01) [G06F 3/0653 (2013.01); G06F 3/0679 (2013.01)] 18 Claims
OG exemplary drawing
 
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.