公开(公告)号：US20230114199A1

公开(公告)日：2023-04-13

申请号：US17731628

申请日：2022-04-28

Applicant: SAMSUNG ELECTRONICS CO., LTD

Inventor： Seung Ryong JANG ,  Chan Ha KIM ,  Gyeong Min NAM

IPC: G06F3/06

Abstract: A storage device including: a nonvolatile memory which includes a first memory block configured to store a first user data; and a controller configured to perform a patrol read on the first user data, wherein the controller is configured to increase a reclaim count for the first user data, when the first user data requires a reclaim, the controller is configured to determine that the first user data is a cold data, when the reclaim count for the first user data is greater than a reclaim threshold value, the controller is configured to perform a first reclaim on the first user data when the first user data is determined to be the cold data, and the nonvolatile memory is configured to store the first user data in a second memory block different from the first memory block in response to the first reclaim.

公开(公告)号：US20230214258A1

公开(公告)日：2023-07-06

申请号：US17949583

申请日：2022-09-21

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Chan Ha KIM ,  Han Kyoo LEE ,  Gyeong Min NAM ,  Seung Ryong JANG

IPC: G06F9/48 ,  G06F9/54 ,  G06K9/62 ,  G06N20/00

CPC classification number: G06F9/4881 ,  G06F9/546 ,  G06K9/6262 ,  G06N20/00

Abstract: A storage controller, including a processor configured to perform a plurality of tasks; and a scheduling module configured to schedule the plurality of tasks through reinforcement learning, and provide a scheduling result to the processor, wherein the scheduling module includes: a resource analysis module configured to analyze a usage history and a usage status of a resource; an access pattern analysis module configured to analyze an access pattern; a queue analysis module configured to analyze information included in a queue; and a performance analysis module configured to analyze a task execution result, wherein the scheduling module is further configured to perform the reinforcement learning using state information and reward information, and wherein the state information and the reward information are determined based on the performing of the plurality of tasks.