WO2013155673A1 - 片内共享缓存的管理方法及装置 - Google Patents

片内共享缓存的管理方法及装置 Download PDF

Info

Publication number
WO2013155673A1
WO2013155673A1 PCT/CN2012/074226 CN2012074226W WO2013155673A1 WO 2013155673 A1 WO2013155673 A1 WO 2013155673A1 CN 2012074226 W CN2012074226 W CN 2012074226W WO 2013155673 A1 WO2013155673 A1 WO 2013155673A1
Authority
WO
WIPO (PCT)
Prior art keywords
storage
user
storage unit
unit
resource
Prior art date
Application number
PCT/CN2012/074226
Other languages
English (en)
French (fr)
Inventor
汪八零
石义军
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to PCT/CN2012/074226 priority Critical patent/WO2013155673A1/zh
Priority to CN201280072414.4A priority patent/CN104246723B/zh
Priority to IN9685DEN2014 priority patent/IN2014DN09685A/en
Priority to EP12874716.9A priority patent/EP2840502A4/en
Publication of WO2013155673A1 publication Critical patent/WO2013155673A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02Addressing or allocation; Relocation
    • G06F12/0223User address space allocation, e.g. contiguous or non contiguous base addressing
    • G06F12/0284Multiple user address space allocation, e.g. using different base addresses
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02Addressing or allocation; Relocation
    • G06F12/0223User address space allocation, e.g. contiguous or non contiguous base addressing
    • G06F12/023Free address space management
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/10Providing a specific technical effect
    • G06F2212/1041Resource optimization
    • G06F2212/1044Space efficiency improvement
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/17Embedded application
    • G06F2212/174Telecommunications system

Definitions

  • the present invention relates to a storage unit sharing technique, and in particular, to an on-chip shared cache management method and apparatus. Background technique
  • SF Spreading Factor
  • the usual processing method is that each user allocates storage space according to the minimum SF, so that when the SF of each user changes, the multipath symbols are guaranteed to be correctly stored.
  • the SF minimum value since all users do not take the SF minimum value, but a distribution of different SF values, if all users allocate buffers according to the minimum SF, the entire storage space will be extremely wasted. Summary of the invention
  • the embodiment of the invention provides a method and a device for managing on-chip shared cache, which can set a suitable shared storage resource pool for multiple users, and realize the requirement of multi-user storage data with less cache space.
  • An on-chip shared cache management method in which a shared storage space is divided into multiple a storage resource pool composed of storage units; a storage table for recording storage units that are free in the storage resource pool; and a resource occupation table for setting a storage unit condition for each user; the method includes:
  • the storage unit When the storage resource is allocated, the storage unit is selected from the storage table for the user, and after the user uses the allocated storage unit, the user reads the corresponding item from the resource occupation table. The storage unit occupied by the user is recorded in the storage table.
  • the method further includes: setting an identifier for each storage unit in the storage resource pool; and recording, in the storage table, identifier information of an idle storage unit;
  • the identifier information of the user occupied storage unit is recorded in the resource occupation table.
  • the resource occupancy table for setting a situation in which each user occupies a storage unit is: setting, for each user, a storage unit of the maximum number of storage units that the user can occupy, and each storage item records a storage unit occupied by the user. Identification information.
  • the size of each storage unit in the storage resource pool is the same, and the size of the storage unit matches the minimum storage requirement of the user.
  • the size of the storage unit matches the minimum storage requirement of the user, and the size of the storage unit is the same as the minimum storage requirement of the user; or the size of the storage unit is greater than the minimum storage requirement of the user.
  • An apparatus for managing an on-chip shared cache comprising a setting unit, an allocating unit, a recycling unit, and a recording unit, wherein:
  • a setting unit configured to divide the shared storage space into a storage resource pool composed of a plurality of storage units; set a storage table that records the storage units that are free in the storage resource pool; and, set a situation in which each user occupies the storage unit Resource occupancy table;
  • An allocating unit configured to select a required storage unit for the user from the storage table when allocating a storage resource to the user;
  • a recovery unit configured to read, after the user finishes using the allocated storage unit, a storage unit occupied by the user from a corresponding item of the user in the resource occupation table;
  • a recording unit configured to record, by the allocating unit, the selected storage unit under the corresponding item of the user in the resource occupation table; and record the storage unit selected by the recycling unit into the storage table.
  • the setting unit is further configured to: set an identifier for each storage unit in the storage resource pool;
  • the recording unit is further configured to: record, in the storage table, identification information of an idle storage unit; and record, in the resource occupation table, identifier information of the user occupied storage unit.
  • the setting unit sets the resource occupation table to set a storage item number of storage units that the user can occupy at most for each user, and each storage item records the identification information of the storage unit occupied by the user.
  • the size of each storage unit in the storage resource pool is the same, and the size of the storage unit matches the minimum storage requirement of the user.
  • the size of the storage unit matches the minimum storage requirement of the user, and the size of the storage unit is the same as the minimum storage requirement of the user; or the size of the storage unit is greater than the minimum storage requirement of the user.
  • the shared storage space is divided into a storage resource pool composed of a plurality of storage units; a storage table for recording the storage units in the storage resource pool is set; and, a resource for each user occupying the storage unit is set. Occupying the table; in this way, when the storage resource needs to be allocated to the user, the required storage unit is selected from the storage table for the user, and the selected storage unit is recorded in the corresponding item of the user in the resource occupation table; The allocated storage unit is recorded in the storage table.
  • the technical solution of the embodiment of the present invention can implement data storage requirements of multiple users by using a cache storage resource pool with a small capacity.
  • Technical solution of the embodiment of the invention It is especially suitable for demodulation of a NodeB baseband physical layer data channel of a WCDMA system.
  • FIG. 1 is a schematic structural diagram of a storage resource pool according to an embodiment of the present invention
  • FIG. 1 is a schematic structural diagram of a storage table according to an embodiment of the present invention.
  • FIG. 3 is a schematic structural diagram of a resource occupation table according to an embodiment of the present invention.
  • FIG. 4 is a schematic structural diagram of a device for managing an on-chip shared cache according to an embodiment of the present invention. detailed description
  • the basic idea of the present invention is: dividing a shared storage space into a storage resource pool composed of a plurality of storage units; setting a storage table for recording storage units that are free in the storage resource pool; and setting a situation in which each user occupies a storage unit.
  • the resource occupancy table in this way, when the storage resource needs to be allocated to the user, the required storage unit is selected from the storage table for the user, and the selected storage unit is recorded in the corresponding item of the user in the resource occupation table; After the allocated storage unit is used, the storage unit occupied by the user is read from the corresponding item of the user in the resource occupation table, and recorded in the storage table.
  • the storage space shared by multiple users is set as a storage resource pool, that is, the storage space shared by multiple users is divided into a plurality of small storage units, and the size of the storage unit matches the minimum storage requirement of the user, and this is The smallest storage unit is called a minimum storage block (MB, Minimum Block).
  • the entire shared storage space is composed of an integer number of MBs.
  • An identifier (ID) is set for each MB, and each MB is distinguished by the MB ID.
  • the size of the storage unit matches the minimum storage requirement of the user as follows: The size of the storage unit is the same as the minimum storage requirement of the user; or the size of the storage unit is greater than the minimum storage requirement of the user.
  • FIG. 1 is a schematic structural diagram of a storage resource pool according to an embodiment of the present invention.
  • a storage resource pool is composed of 512 MBs as an example.
  • MB0 to MB511 are respectively shown in the table. Indicates the standard i of the MB.
  • the storage table is implemented in a first in first out (FIFO) storage manner.
  • FIFO first in first out
  • the storage table stores all MB IDs of the entire storage resource pool, that is, in the initial state, all the storage resource pools are in the storage pool.
  • MB is not occupied by users and is available.
  • FIG. 1 is a schematic structural diagram of a storage table according to an embodiment of the present invention. As shown in FIG. 1 , it is a schematic diagram of storing MB in a storage table in an initial state.
  • the resource occupancy table needs to be set for each user, and the number of each user entry in the resource occupation table is the same.
  • the maximum number of users in the resource occupation table depends on the maximum number of MBs that the user can occupy in the application scenario. Make settings, each entry is used to store the MB allocated by the user.
  • FIG. 3 is a schematic structural diagram of a resource occupation table according to an embodiment of the present invention.
  • a resource occupation table has 256 users, and each user can occupy up to 16 MBs, but corresponding to an application scenario, each The MB occupied by users is different.
  • the specified MB in the table is stored in the table, and then the corresponding MB ID of all the MBs allocated to the user is stored in the entry corresponding to the user in the resource occupation table. Achieve the allocation of MB.
  • the function of multi-user shared resource pool can be achieved by allocating and reclaiming MB.
  • the technical solution of the embodiment of the present invention is particularly applicable to demodulation of a NodeB baseband physical layer data channel of a WCDMA system, as applicable to an intermediate result of a plurality of chips accumulated in the foregoing cache multipath demodulation process.
  • the essence of the technical solution of the present invention will be further clarified by specific examples below.
  • a total of 256 users need to share the cache area, wherein each user uses up to 16 MB, that is, 512 MB of the buffer area is shared by 256 users.
  • the MB that is currently in the idle state is recorded.
  • a certain amount of MB is selected from the storage table, and when the MB used by the user is deleted, the All MB resources allocated by the user are recorded in the storage table.
  • the storage table can be implemented by the FIFO memory. Taking the storage table shown in Fig. 2 as an example, a FIFO with a depth of 512 and a width of 9 bits can be realized. Each unit of the FIFO memory can record the MB ID of an idle MB. The storage table is full in the initialization state, indicating that there are 512 free MB resources available at the beginning. When the MB resource is allocated, the corresponding MB ID is read from the FIFO. When the MB resource is reclaimed, the recovered MB ID is recorded in the storage table.
  • the resource occupancy table records the MB resources occupied by each user in the working process. You can obtain the MB resources currently used by each user from the resource occupancy table.
  • the MB in the storage entry of the user is read from the resource occupation table according to the MB resource recovery command, and the IDs of the MBs are recorded in the storage table, thereby completing the user MB resource. Recycling.
  • the technical solution of the embodiment of the present invention can implement data storage requirements of multiple users by using a cache storage resource pool with a small capacity.
  • the technical solution of the embodiment of the present invention is particularly applicable to demodulation of a NodeB baseband physical layer data channel of a WCDMA system.
  • the device for managing an on-chip shared cache of this embodiment includes a setting unit 40, an allocating unit 41, and a reclaiming unit 42. Recording unit 43, wherein:
  • the setting unit 40 is configured to divide the shared storage space into a storage resource pool composed of a plurality of storage units; set a storage table that records the storage units that are free in the storage resource pool; and, set a storage unit condition for each user.
  • Resource occupancy table ;
  • the allocating unit 41 is configured to: when the storage resource is allocated to the user, select a required storage unit for the user from the storage table;
  • the recycling unit 42 is configured to read, after the user finishes using the allocated storage unit, the storage unit occupied by the user from the corresponding item of the user in the resource occupation table;
  • a recording unit 43 configured to record, by the allocating unit, the selected storage unit in a corresponding item of the user in the resource occupation table; and record the storage unit selected by the recycling unit into the storage table .
  • the setting unit 40 is further configured to: set an identifier for each storage unit in the storage resource pool; the recording unit 43 is further configured to: record, in the storage table, identifier information of an idle storage unit; Recording the identification information of the user occupying the storage unit.
  • the setting unit 40 sets the resource occupation table to set a storage item number of storage units that the user can occupy at most for each user, and each storage item records one identification information of the storage unit occupied by the user.
  • the storage units in the storage resource pool have the same size, and the size of the storage unit matches the minimum storage requirement of the user.
  • the size of the storage unit matches the minimum storage requirement of the user, where: the size of the storage unit is the same as the minimum storage requirement of the user; or the size of the storage unit is greater than the minimum storage requirement of the user.
  • the foregoing storage table can be implemented by a FIFO memory, and the foregoing resource occupancy table can be implemented by a dual port RAM.
  • the technical solution of the embodiment of the present invention can implement data storage requirements of multiple users by using a cache storage resource pool with a small capacity.
  • the technical solution of the embodiment of the present invention is particularly applicable to demodulation of a NodeB baseband physical layer data channel of a WCDMA system.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Memory System Of A Hierarchy Structure (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Memory System (AREA)

Abstract

本发明涉及缓存共享技术,公开了一种片内共享缓存的管理方法及装置,用于片内的缓存共享,所述方法包括:将共享的存储空间划分为由多个存储单元组成的存储资源池;设置记录所述存储资源池中空闲的存储单元的存储表;以及,设置每个用户占用存储单元情况的资源占用表;分配存储资源时,从所述存储表中为所述用户选取所需的存储单元,并将所选取的存储单元记录于所述资源占用表中所述用户的对应项下;所述用户使用完所分配的存储单元后,从所述资源占用表中该用户的对应项下读取所述用户占用的存储单元,并记录到所述存储表中。本发明实施例的技术方案,以容量较小的缓存存储资源池,可以实现多个用户的数据存储需求。

Description

片内共享緩存的管理方法及装置 技术领域
本发明涉及存储单元共享技术, 特别涉及一种片内共享緩存的管理方 法及装置。 背景技术
在宽带码分多址( WCDMA, Wideband Code Division Multiple Access ) 的基站( NodeB )基带物理层数据信道解调过程中, 需要緩存多径解调过程 中多个码片累加后的中间结果, 由于不同用户的数据信道扩频因子 (SF, Spreading Factor )可能是不同的, 每个 SF数目的码片累加即得出一个多径 符号, 每个符号需要独立存储。 因而在相同的时间段内不同用户存储的中 间多径符号的个数也不同。
针对这种情况,通常的处理方法是每个用户都按照最小的 SF分配存储 空间, 这样可以保证各个用户的 SF的发生变化时, 都能保证其多径符号被 正确存储。但是实际应用场景下,由于不会出现所有的用户都取 SF最小值, 而是一个不同 SF取值的分布,如果所有的用户都按照最小的 SF分配緩存, 势必导致整个存储空间极大浪费。 发明内容
本发明实施例提供一种片内共享緩存的管理方法及装置, 能为多个用 户设置合适的共享存储资源池, 以较少的緩存空间实现多用户存储数据的 需求。
本发明实施例提供的具体技术方案如下:
一种片内共享緩存的管理方法, 其中, 将共享的存储空间划分为由多 个存储单元组成的存储资源池; 设置记录所述存储资源池中空闲的存储单 元的存储表; 以及, 设置每个用户占用存储单元情况的资源占用表; 所述 方法包括:
分配存储资源时, 从所述存储表中为所述用户选取所需的存储单元, 所述用户使用完所分配的存储单元后, 从所述资源占用表中该用户的 对应项下读取所述用户占用的存储单元, 并记录到所述存储表中。
其中, 所述方法还包括: 为所述存储资源池中的各存储单元设置标识; 所述存储表中记录空闲的存储单元的标识信息;
所述资源占用表中记录所述用户占用存储单元的标识信息。
其中, 所述设置每个用户占用存储单元情况的资源占用表, 为: 为每个用户设置该用户最多能占用的存储单元数量个存储项, 每个存 储项记录一个该用户所占用的存储单元的标识信息。
其中, 所述存储资源池中各存储单元的大小相同, 且所述存储单元的 大小与用户的最小存储需求相匹配。
其中, 所述存储单元的大小与用户的最小存储需求相匹配, 为: 所述存储单元的大小与用户的最小存储需求相同; 或者, 所述存储单 元的大小大于用户的最小存储需求。
一种片内共享緩存的管理装置, 所述装置包括设置单元、 分配单元、 回收单元和记录单元, 其中:
设置单元, 用于将共享的存储空间划分为由多个存储单元组成的存储 资源池; 设置记录所述存储资源池中空闲的存储单元的存储表; 以及, 设 置每个用户占用存储单元情况的资源占用表;
分配单元, 用于在为用户分配存储资源时, 从所述存储表中为所述用 户选取所需的存储单元; 回收单元, 用于在所述用户使用完所分配的存储单元后, 从所述资源 占用表中该用户的对应项下读取所述用户占用的存储单元;
记录单元, 用于所述分配单元将所选取的存储单元记录于所述资源占 用表中所述用户的对应项下; 以及, 将所述回收单元选取的存储单元记录 到所述存储表中。
其中, 所述设置单元还用于, 为所述存储资源池中的各存储单元设置 标识;
所述记录单元还用于, 在所述存储表中记录空闲的存储单元的标识信 息; 在所述资源占用表中记录所述用户占用存储单元的标识信息。
其中, 所述设置单元设置资源占用表为, 为每个用户设置该用户最多 能占用的存储单元数量个存储项, 每个存储项记录一个该用户所占用的存 储单元的标识信息。
其中, 所述存储资源池中各存储单元的大小相同, 且所述存储单元的 大小与用户的最小存储需求相匹配。
其中, 所述存储单元的大小与用户的最小存储需求相匹配, 为: 所述存储单元的大小与用户的最小存储需求相同; 或者, 所述存储单 元的大小大于用户的最小存储需求。
本发明实施例中, 将共享的存储空间划分为由多个存储单元组成的存 储资源池; 设置记录存储资源池中空闲的存储单元的存储表; 并且, 设置 每个用户占用存储单元情况的资源占用表; 这样, 当需要为用户分配存储 资源时, 从存储表中为用户选取所需的存储单元, 并将所选取的存储单元 记录于资源占用表中所述用户的对应项下; 用户使用完所分配的存储单元 记录到存储表中。 这样, 本发明实施例的技术方案, 以容量较小的緩存存 储资源池, 可以实现多个用户的数据存储需求。 本发明实施例的技术方案 尤其适用于 WCDMA系统的 NodeB基带物理层数据信道的解调。 附图说明
图 1为本发明实施例的存储资源池的结构示意图;
图 1为本发明实施例的存储表的结构示意图;
图 3为本发明实施例的资源占用表的结构示意图;
图 4为本发明实施例的片内共享緩存的管理装置的组成结构示意图。 具体实施方式
本发明的基本思想为: 将共享的存储空间划分为由多个存储单元组成 的存储资源池; 设置记录存储资源池中空闲的存储单元的存储表; 并且, 设置每个用户占用存储单元情况的资源占用表; 这样, 当需要为用户分配 存储资源时, 从存储表中为用户选取所需的存储单元, 并将所选取的存储 单元记录于资源占用表中所述用户的对应项下; 用户使用完所分配的存储 单元后, 从资源占用表中该用户的对应项下读取所述用户占用的存储单元, 并记录到存储表中。
下面结合附图对本发明优选的实施方式进行详细说明。
本发明中, 将多用户共享的存储空间设定为存储资源池, 即将多用户 共享的存储空间划分成多个小的存储单元, 该存储单元的大小和用户的最 小存储需求相匹配, 将这个最小的存储单元称为一个最小存储块(MB, Minimum Block ), 整个共享存储空间由整数个 MB构成, 为各个 MB设置 标识(ID ), 以 MB ID区分各个 MB。 其中, 存储单元的大小与用户的最小 存储需求相匹配为: 存储单元的大小与用户的最小存储需求相同; 或者, 存储单元的大小大于用户的最小存储需求。
图 1为本发明实施例的存储资源池的结构示意图, 如图 1所示, 以存 储资源池由 512个 MB组成为例进行的示意, 图中, MB0至 MB511分别表 示 MB的标 i只。
本发明中,还需设置一个存储表,用于存储没有被所有用户占用的 MB。 该存储表以先进先出 (FIFO, First In First Out )存储方式实现, 初始状态 下, 该存储表存储了整个存储资源池所有 MB ID, 也就是说, 初始状态下, 存储资源池中的所有 MB都是未被用户占用的, 都是可用的。
图 1为本发明实施例的存储表的结构示意图, 如图 1所示, 为初始状 态下存储表中存储 MB的示意图。
本发明中, 还需为每个用户设置资源占用表, 资源占用表中每个用户 表项数目相同, 资源占用表中用户的最大表项数目依据应用场景下用户能 够需要占用的最大 MB个数进行设置, 每个表项用于存储该用户所分配的 MB。
图 3为本发明实施例的资源占用表的结构示意图, 如图 3所示, 资源 占用表中一种有 256个用户, 每个用户最多可以占用 16个 MB, 但是对应 一种应用场景, 每个用户占用的 MB不同。
当一个用户需要占用存储资源时,先存储表中申请指定个数的 MB, 然 后将为该用户分配到的所有 MB的对应 MB ID存储到资源占用表中该用户 所对应的表项中, 从而实现 MB的分配。
当一个用户使用完所分配的 MB后, 先从资源占用表中该用户对应的 表项中读取为该用户分配的所有 MB的 MB ID, 并将所读取的 MB ID记录 到存储表中, 从而实现 MB的回收。
本发明中, 通过对 MB 的分配和回收, 即能够达到多用户共享资源池 的功能。
本发明实施例的技术方案尤其适用于 WCDMA系统的 NodeB基带物理 层数据信道的解调, 如可适用于前述緩存多径解调过程中多个码片累加后 的中间结果。 以下通过具体示例, 进一步阐明本发明技术方案的实质。
设有一个緩存区总共包含 512个最小存储块 MB,—个最小存储块是该 緩存区最小的粒度单元, 每个用户使用緩沖区时, 占用的緩存资源的大小 为最小存储块的整数倍, 该緩存区的结构如图 1所示。
另设共有 256个用户需要共享该緩存区, 其中每个用户最多使用 16个 MB, 也就是说, 512个 MB的緩存区被 256个用户共享。 以下介绍该 256 用户共享緩存区的管理方法。
图 1中, 总共有 512个 MB, 为每个 MB设置一个 MB ID, 如图 1所 示, MB 0的 MB ID=0, MB 1的 MB ID=1 ,以此类推, MB 511的 MB ID=511。
如图 2所示,存储表中记录该緩沖区当前处于空闲状态的 MB, 为新建 用户分配 MB 时, 要从存储表中选取一定量的 MB, 而删除用户所使用的 MB时, 将为该用户所分配的所有 MB资源记录到存储表中。
存储表可以由 FIFO存储器实现, 以图 2所示的存储表为例, 深度为 512、 宽度为 9比特的 FIFO即可实现。 FIFO存储器的每个单元可记录一个 空闲 MB的 MB ID。 存储表初始化状态下是满的, 表示开始有 512个空闲 MB资源可用。 分配 MB资源时, 对应从 FIFO读出相应的 MB ID, 回收 MB资源时 , 对应往存储表中记录该回收的 MB ID。
资源占用表中记录工作过程中各用户的占用 MB资源的情况, 可以从 资源占用表中获取每个用户当前用的 MB资源。
用户的资源占用表可以由双口 RAM实现,以图 3所示的资源占用表为 例, 选用深度为 256 X 16=4096、 宽度为 10比特的双口 RAM即可实现。 每 个单元可记录给工作中的用户分配的某个 MB资源的 MB ID。 资源占用表 中, 以用户 ID为索引对应该用户的存储表项, 每个用户的表项占用 16个 连续地址单元。 为用户分配 MB资源时, 将所分配的 MBID记录于资源占 用表中该用户存储表项的对应位置即可。 当需要为用户分配 MB资源时, 从图 2所示的存储表中选取 MB资源 命令中所要求的 MB, 并将所选取的 MB的 ID记录于资源占用表中该用户 存储表项的对应位置, 从而完成用户的 MB资源的分配。
某用户没有业务而不需要 MB时, 依据 MB资源回收命令, 从资源占 用表中读取该用户的存储表项中的 MB, 将这些 MB的 ID记录于存储表中 中, 从而完成用户 MB资源的回收。
本发明实施例的技术方案, 以容量较小的緩存存储资源池, 可以实现 多个用户的数据存储需求。 本发明实施例的技术方案尤其适用于 WCDMA 系统的 NodeB基带物理层数据信道的解调。
图 4为本发明实施例的片内共享緩存的管理装置的组成结构示意图, 如图 4所示, 本实施例的片内共享緩存的管理装置包括设置单元 40、 分配 单元 41、 回收单元 42和记录单元 43 , 其中:
设置单元 40, 用于将共享的存储空间划分为由多个存储单元组成的存 储资源池; 设置记录所述存储资源池中空闲的存储单元的存储表; 以及, 设置每个用户占用存储单元情况的资源占用表;
分配单元 41 , 用于在为用户分配存储资源时, 从所述存储表中为所述 用户选取所需的存储单元;
回收单元 42, 用于在所述用户使用完所分配的存储单元后, 从所述资 源占用表中该用户的对应项下读取所述用户占用的存储单元;
记录单元 43 , 用于所述分配单元将所选取的存储单元记录于所述资源 占用表中所述用户的对应项下; 以及, 将所述回收单元选取的存储单元记 录到所述存储表中。
上述设置单元 40还用于,为所述存储资源池中的各存储单元设置标识; 记录单元 43还用于,在所述存储表中记录空闲的存储单元的标识信息; 在所述资源占用表中记录所述用户占用存储单元的标识信息。 上述设置单元 40设置资源占用表为, 为每个用户设置该用户最多能占 用的存储单元数量个存储项, 每个存储项记录一个该用户所占用的存储单 元的标识信息。
上述存储资源池中各存储单元的大小相同, 且所述存储单元的大小与 用户的最小存储需求相匹配。 其中, 所述存储单元的大小与用户的最小存 储需求相匹配, 为: 所述存储单元的大小与用户的最小存储需求相同; 或 者, 所述存储单元的大小大于用户的最小存储需求。
前述存储表可由 FIFO存储器实现, 前述资源占用表可由双口 RAM实 现。
本领域技术人员应当理解, 图 4 中所示的片内共享緩存的管理装置中 的各处理单元的实现功能可参照前述片内共享緩存的管理方法的相关描述 而理解。 本领域技术人员应当理解, 图 4所示的片内共享緩存的管理装置 中各处理单元的功能可通过运行于处理器上的程序而实现, 也可通过具体 的逻辑电路而实现。
以上所述, 仅为本发明的较佳实施例而已, 并非用于限定本发明的保 护范围。
工业实用性
本发明实施例的技术方案, 以容量较小的緩存存储资源池, 可以实现 多个用户的数据存储需求。 本发明实施例的技术方案尤其适用于 WCDMA 系统的 NodeB基带物理层数据信道的解调。

Claims

权利要求书
1、 一种片内共享緩存的管理方法, 其中, 将共享的存储空间划分为由 多个存储单元组成的存储资源池; 设置记录所述存储资源池中空闲的存储 单元的存储表; 以及, 设置每个用户占用存储单元情况的资源占用表; 所 述方法包括:
分配存储资源时, 从所述存储表中为所述用户选取所需的存储单元, 并将所选取的存储单元记录于所述资源占用表中所述用户的对应项下; 所述用户使用完所分配的存储单元后, 从所述资源占用表中该用户的 对应项下读取所述用户占用的存储单元, 并记录到所述存储表中。
2、 根据权利要求 1所述的方法, 其中, 所述方法还包括: 为所述存储 资源池中的各存储单元设置标识;
所述存储表中记录空闲的存储单元的标识信息;
所述资源占用表中记录所述用户占用存储单元的标识信息。
3、 根据权利要求 2所述的方法, 其中, 所述设置每个用户占用存储单 元情况的资源占用表, 为:
为每个用户设置该用户最多能占用的存储单元数量个存储项, 每个存 储项记录一个该用户所占用的存储单元的标识信息。
4、 根据权利要求 1至 3任一项所述的方法, 其中, 所述存储资源池中 各存储单元的大小相同, 且所述存储单元的大小与用户的最小存储需求相 匹配。
5、 根据权利要求 4所述的方法, 其中, 所述存储单元的大小与用户的 最小存储需求相匹配, 为:
所述存储单元的大小与用户的最小存储需求相同; 或者, 所述存储单 元的大小大于用户的最小存储需求。
6、 一种片内共享緩存的管理装置, 其中, 所述装置包括设置单元、 分 配单元、 回收单元和记录单元, 其中:
设置单元, 用于将共享的存储空间划分为由多个存储单元组成的存储 资源池; 设置记录所述存储资源池中空闲的存储单元的存储表; 以及, 设 置每个用户占用存储单元情况的资源占用表;
分配单元, 用于在为用户分配存储资源时, 从所述存储表中为所述用 户选取所需的存储单元;
回收单元, 用于在所述用户使用完所分配的存储单元后, 从所述资源 占用表中该用户的对应项下读取所述用户占用的存储单元;
记录单元, 用于所述分配单元将所选取的存储单元记录于所述资源占 用表中所述用户的对应项下; 以及, 将所述回收单元选取的存储单元记录 到所述存储表中。
7、 根据权利要求 6所述的装置, 其中, 所述设置单元还用于, 为所述 存储资源池中的各存储单元设置标识;
所述记录单元还用于, 在所述存储表中记录空闲的存储单元的标识信 息; 在所述资源占用表中记录所述用户占用存储单元的标识信息。
8、 根据权利要求 7所述的装置, 其中, 所述设置单元设置资源占用表 为, 为每个用户设置该用户最多能占用的存储单元数量个存储项, 每个存 储项记录一个该用户所占用的存储单元的标识信息。
9、 根据权利要求 6至 8任一项所述的装置, 其中, 所述存储资源池中 各存储单元的大小相同, 且所述存储单元的大小与用户的最小存储需求相 匹配。
10、 根据权利要求 9所述的装置, 其中, 所述存储单元的大小与用户 的最小存储需求相匹配, 为:
所述存储单元的大小与用户的最小存储需求相同; 或者, 所述存储单 元的大小大于用户的最小存储需求。
PCT/CN2012/074226 2012-04-17 2012-04-17 片内共享缓存的管理方法及装置 WO2013155673A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
PCT/CN2012/074226 WO2013155673A1 (zh) 2012-04-17 2012-04-17 片内共享缓存的管理方法及装置
CN201280072414.4A CN104246723B (zh) 2012-04-17 2012-04-17 片内共享缓存的管理方法及装置
IN9685DEN2014 IN2014DN09685A (zh) 2012-04-17 2012-04-17
EP12874716.9A EP2840502A4 (en) 2012-04-17 2012-04-17 ADMINISTRATIVE PROCESS AND DEVICE FOR COMMON ON-CHIP INTERMEDIATE STORAGE

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2012/074226 WO2013155673A1 (zh) 2012-04-17 2012-04-17 片内共享缓存的管理方法及装置

Publications (1)

Publication Number Publication Date
WO2013155673A1 true WO2013155673A1 (zh) 2013-10-24

Family

ID=49382797

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2012/074226 WO2013155673A1 (zh) 2012-04-17 2012-04-17 片内共享缓存的管理方法及装置

Country Status (4)

Country Link
EP (1) EP2840502A4 (zh)
CN (1) CN104246723B (zh)
IN (1) IN2014DN09685A (zh)
WO (1) WO2013155673A1 (zh)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016097813A1 (en) * 2014-12-14 2016-06-23 Via Alliance Semiconductor Co., Ltd. Set associative cache memory with heterogeneous replacement policy
US9652398B2 (en) 2014-12-14 2017-05-16 Via Alliance Semiconductor Co., Ltd. Cache replacement policy that considers memory access type
US9652400B2 (en) 2014-12-14 2017-05-16 Via Alliance Semiconductor Co., Ltd. Fully associative cache memory budgeted by memory access type
US9898411B2 (en) 2014-12-14 2018-02-20 Via Alliance Semiconductor Co., Ltd. Cache memory budgeted by chunks based on memory access type
US9910785B2 (en) 2014-12-14 2018-03-06 Via Alliance Semiconductor Co., Ltd Cache memory budgeted by ways based on memory access type
CN108737476A (zh) * 2017-04-21 2018-11-02 杭州海康威视数字技术股份有限公司 云存储系统、媒体数据存储方法及系统
CN109862591A (zh) * 2019-02-28 2019-06-07 电子科技大学 一种基于Qos空口切片的带宽借用与缓存共享方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112311610B (zh) * 2020-12-25 2021-05-18 视联动力信息技术股份有限公司 一种非ip体系下实现qos保障的通信方法和装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1529242A (zh) * 2003-09-29 2004-09-15 港湾网络有限公司 路由器流量管理芯片缓存管理的实现方法
CN1532708A (zh) * 2003-03-19 2004-09-29 华为技术有限公司 一种静态内存管理方法
CN1819544A (zh) * 2005-01-05 2006-08-16 华为技术有限公司 一种基于bitmap表的缓存管理方法
CN1855881A (zh) * 2005-04-28 2006-11-01 华为技术有限公司 动态共享存储器存储空间的实现方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2009098547A1 (en) * 2008-02-08 2009-08-13 Freescale Semiconductor, Inc. Memory management
US8271760B2 (en) * 2009-06-04 2012-09-18 Petit Brent J Spatial locality of file system block allocations for related items
US20110246742A1 (en) * 2010-04-01 2011-10-06 Kogen Clark C Memory pooling in segmented memory architecture

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1532708A (zh) * 2003-03-19 2004-09-29 华为技术有限公司 一种静态内存管理方法
CN1529242A (zh) * 2003-09-29 2004-09-15 港湾网络有限公司 路由器流量管理芯片缓存管理的实现方法
CN1819544A (zh) * 2005-01-05 2006-08-16 华为技术有限公司 一种基于bitmap表的缓存管理方法
CN1855881A (zh) * 2005-04-28 2006-11-01 华为技术有限公司 动态共享存储器存储空间的实现方法

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016097813A1 (en) * 2014-12-14 2016-06-23 Via Alliance Semiconductor Co., Ltd. Set associative cache memory with heterogeneous replacement policy
TWI559143B (zh) * 2014-12-14 2016-11-21 上海兆芯集成電路有限公司 具多樣替換策略的集合關聯快取記憶體
US9652398B2 (en) 2014-12-14 2017-05-16 Via Alliance Semiconductor Co., Ltd. Cache replacement policy that considers memory access type
US9652400B2 (en) 2014-12-14 2017-05-16 Via Alliance Semiconductor Co., Ltd. Fully associative cache memory budgeted by memory access type
US9811468B2 (en) 2014-12-14 2017-11-07 Via Alliance Semiconductor Co., Ltd. Set associative cache memory with heterogeneous replacement policy
US9898411B2 (en) 2014-12-14 2018-02-20 Via Alliance Semiconductor Co., Ltd. Cache memory budgeted by chunks based on memory access type
US9910785B2 (en) 2014-12-14 2018-03-06 Via Alliance Semiconductor Co., Ltd Cache memory budgeted by ways based on memory access type
CN108737476A (zh) * 2017-04-21 2018-11-02 杭州海康威视数字技术股份有限公司 云存储系统、媒体数据存储方法及系统
CN109862591A (zh) * 2019-02-28 2019-06-07 电子科技大学 一种基于Qos空口切片的带宽借用与缓存共享方法
CN109862591B (zh) * 2019-02-28 2021-04-13 电子科技大学 一种基于Qos空口切片的带宽借用与缓存共享方法

Also Published As

Publication number Publication date
EP2840502A1 (en) 2015-02-25
IN2014DN09685A (zh) 2015-07-31
CN104246723B (zh) 2018-08-10
EP2840502A4 (en) 2015-06-03
CN104246723A (zh) 2014-12-24

Similar Documents

Publication Publication Date Title
WO2013155673A1 (zh) 片内共享缓存的管理方法及装置
CN102193872B (zh) 存储系统
CN105677242B (zh) 冷热数据的分离方法和装置
US7733892B2 (en) Buffer management method based on a bitmap table
CN104090847B (zh) 一种固态存储设备的地址分配方法
CN106502806B (zh) 一种总线协议命令处理装置及相关方法
US10437519B2 (en) Method and mobile terminal for processing write request
CN108108128A (zh) 一种混合ssd的磨损均衡方法及ssd
KR101687762B1 (ko) 저장 장치 및 그것의 커맨드 스케줄링 방법
US9223373B2 (en) Power arbitration for storage devices
US9047174B2 (en) Method and system for hierarchically managing storage resources
CN103713861A (zh) 一种基于层次划分的文件处理方法及系统
CN103425435A (zh) 磁盘存储方法及磁盘存储系统
CN103874987B (zh) 嵌入式多媒体卡分区存储空间调整方法和终端
US20130061009A1 (en) High Performance Free Buffer Allocation and Deallocation
WO2020155552A1 (zh) 一种基于地址映射的单盘多用户物理隔离方法
CN103116475B (zh) 一种自动精简配置扩容的方法
CN108205501B (zh) 内存回收方法及装置、计算机装置及计算机可读存储介质
CN104991737A (zh) 一种基于存储卡阵列架构的硬盘实现方法
CN108287760B (zh) 终端设备控制方法及装置、终端设备及计算机可读存储介质
CN107145450B (zh) 视频监控数据存储方法和装置
CN104503924B (zh) 一种分层存储系统中的区域分配方法及装置
CN106502591B (zh) 一种基于混合存储产品的存储控制方法和存储控制装置
CN105224258B (zh) 一种数据缓冲区的复用方法与系统
KR102063281B1 (ko) 파일에 접근하기 위한 방법과 장치, 및 저장 시스템

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12874716

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2012874716

Country of ref document: EP