知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

9 records (took 0.019 seconds)

    Method and apparatus for managing memory
    1.
    发明授权
    Method and apparatus for managing memory 有权

    公开(公告)号：US10409498B2

    公开(公告)日：2019-09-10

    申请号：US15172951

    申请日：2016-06-03

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Kibeom Kim ,  Jinkyu Koo ,  Kwonsik Kim ,  Hyojeong Lee

    IPC: G06F3/06

    Abstract: A method for managing a memory and an electronic device are provided. The method includes setting a list including an exclusive relationship, between a page unit memory allocation requester and a contiguous memory allocation requester, corresponding to a segment of a contiguous memory allocation region, receiving a memory allocation request, confirming whether the memory allocation requester comprises the memory allocation requester included in the list, and allocating a page of the contiguous memory allocation region corresponding to the segment, if the memory allocation requester is included in the list.

    Electronic device and memory allocation method therefor
    2.
    发明授权
    Electronic device and memory allocation method therefor 有权

    公开(公告)号：US10255175B2

    公开(公告)日：2019-04-09

    申请号：US15570047

    申请日：2016-05-11

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Kwonsik Kim ,  Jihwan Park ,  Kibeom Kim ,  Hyojeong Lee

    IPC: G06F9/52 ,  G06F12/02 ,  G06F12/08 ,  G06F1/32 ,  G06F3/06

    Abstract: An electronic device according to an embodiment of the present disclosure comprises: a memory including a plurality of pages; and a controller selecting a page area having no non-movable page from a plurality of page areas in the memory based on page state information on the plurality of page areas if a memory defragmentation request occurs when a memory allocation request is made, moving a movable page in the selected page area, and allocating the selected page area in response to the memory allocation request after the movable page has been moved. However, the present disclosure is not limited to the above embodiment, and other embodiments are possible.

    Resource allocation method and apparatus of GPU
    3.
    发明授权
    Resource allocation method and apparatus of GPU 有权
    GPU的资源分配方法和装置

    公开(公告)号：US09454401B2

    公开(公告)日：2016-09-27

    申请号：US13750837

    申请日：2013-01-25

    Applicant: Samsung Electronics Co., Ltd

    Inventor： Kwonsik Kim ,  Youngwoo Ahn ,  Jeongig Song ,  Inchoon Yeo

    IPC: G06F9/46 ,  G06F9/50 ,  G06T1/20

    CPC classification number: G06F9/5011 ,  G06F9/5066 ,  G06T1/20

    Abstract: A resource allocation method and apparatus utilize the GPU resource efficiently by sorting the tasks using General Purpose GPU (GPGPU) into operations and combining the same operations into a request. The resource allocation method of a Graphic Processing Unit (GPU) according to the present disclosure includes receiving a task including at least one operation; storing the at least one operation in unit of request; merging data of same operations per request; and allocating GPU resource according to an execution order the request.

    Abstract translation: 资源分配方法和装置通过使用通用GPU（GPGPU）将任务排序到操作中并且将相同的操作组合成请求来有效地利用GPU资源。 根据本公开的图形处理单元（GPU）的资源分配方法包括：接收包括至少一个操作的任务; 以请求为单位存储所述至少一个操作; 合并每个请求相同操作的数据; 并根据请求的执行顺序分配GPU资源。

    RESOURCE ALLOCATION METHOD AND APPARATUS OF GPU
    6.
    发明申请
    RESOURCE ALLOCATION METHOD AND APPARATUS OF GPU 有权
    资源分配方法和GPU设备

    公开(公告)号：US20130198757A1

    公开(公告)日：2013-08-01

    申请号：US13750837

    申请日：2013-01-25

    Applicant: Samsung Electronics Co., Ltd.

    Inventor： Kwonsik Kim ,  Youngwoo Ahn ,  Jeongig Song ,  Inchoon Yeo

    IPC: G06F9/50

    CPC classification number: G06F9/5011 ,  G06F9/5066 ,  G06T1/20

    Abstract: A resource allocation method and apparatus utilize the GPU resource efficiently by sorting the tasks using General Purpose GPU (GPGPU) into operations and combining the same operations into a request. The resource allocation method of a Graphic Processing Unit (GPU) according to the present disclosure includes receiving a task including at least one operation; storing the at least one operation in unit of request; merging data of same operations per request; and allocating GPU resource according to an execution order the request.

    Abstract translation: 资源分配方法和装置通过使用通用GPU（GPGPU）将任务排序到操作中并将相同的操作组合成请求来有效地利用GPU资源。 根据本公开的图形处理单元（GPU）的资源分配方法包括接收包括至少一个操作的任务; 以请求为单位存储所述至少一个操作; 合并每个请求相同操作的数据; 并根据请求的执行顺序分配GPU资源。

Patent Agency Ranking