CN104133774A - Method of managing non-volatile memory and non-volatile storage device thereof - Google Patents

Method of managing non-volatile memory and non-volatile storage device thereof Download PDF

Info

Publication number
CN104133774A
CN104133774A CN 201310547079 CN201310547079A CN104133774A CN 104133774 A CN104133774 A CN 104133774A CN 201310547079 CN201310547079 CN 201310547079 CN 201310547079 A CN201310547079 A CN 201310547079A CN 104133774 A CN104133774 A CN 104133774A
Authority
CN
China
Prior art keywords
memory
data
page
plurality
nonvolatile memory
Prior art date
Application number
CN 201310547079
Other languages
Chinese (zh)
Inventor
戴铭佑
刘亦峻
Original Assignee
擎泰科技股份有限公司
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
Priority to US201361818884P priority Critical
Priority to US13/953,764 priority patent/US20140328127A1/en
Application filed by 擎泰科技股份有限公司 filed Critical 擎泰科技股份有限公司
Publication of CN104133774A publication Critical patent/CN104133774A/en

Links

Abstract

The invention discloses a method of managing a non-volatile memory and a non-volatile storage device thereof. The non-volatile memory comprises a plurality of memory blocks; and each of the plurality of memory blocks includes a plurality of memory pages. The method includes the steps of partitioning a memory page among the plurality of memory pages into a plurality of clusters; and writing data and a mapping information corresponding to the data into different clusters of the plurality of clusters.

Description

管理非易失性存储器的方法及其非易失性存储装置 The method of managing a nonvolatile memory and the nonvolatile memory device

技术领域 FIELD

[0001] 本发明涉及一种管理非易失性存储器的方法及其非易失性存储装置,尤其涉及一种可在一次写入一存储页的操作中同时写入数据及映射信息的管理非易失性存储器的方法及其非易失性存储装置。 [0001] The present invention relates to a method of managing a nonvolatile memory and the nonvolatile memory device, particularly to a data and mapping information may be simultaneously written in a write-once memory page management in the non-operation the method of a volatile memory and a nonvolatile memory device.

背景技术 Background technique

[0002] 对一般非易失性存储装置而言,主机通常是以区段(Sector)为单位对非易失性存储器进行管理,而非易失性存储器的最小写入单位则是根据一存储页(Memory Page)的大小而定。 [0002] For a general nonvolatile memory device, usually a host section (Sector) are managed as a unit of the nonvolatile memory, the minimum unit of writing in the nonvolatile memory is a memory in accordance with page (Memory page) in size. 传统上,会将一区段的大小设计为相等于一存储页的大小,使得页映射时的写入效率可达到最大。 Traditionally, the size of a section will be designed to be equal to the size of a memory page, so that the write efficiency can be maximized page mapping. 当主机欲将一区段大小的数据写入非易失性存储器时,此数据可通过页映射的方式写入一完整的存储页,而不浪费存储页中的任何空间。 When a host wishing sector size data written in the nonvolatile memory, the data can be written to a memory page by full page mapping manner, without wasting any space in the memory page.

[0003] 近年来,随着人们对存储空间的高度需求,非易失性存储器的大小不断地增加,而存储页的大小也随之而增加,然而,区段的大小却没有明显变化。 [0003] In recent years, as people's high demand for storage space, the size of the non-volatile memory is increasing, while the size of the memory page also will increase, however, the size of the sector did not change significantly. 在此情况下,当一存储页中一区段大小的数据欲更新时,此区段的更新数据必须写入一新的存储页,而在新的存储页中其它多余的记忆空间则必须填补原存储页中其它数据,因而造成存储器空间的浪费以及写入效率的低落。 In this case, when a section of a memory page size of data to be updated, to update data in a segment to be written to the new memory page, and the page in the new storage space in other extra memory must be filled other data stored in the original page, resulting in a waste of memory space and writing efficiency is low. 举例来说,请参考图1,图1为用户数据Dl写入一存储页Pl的示意图。 For example, please refer to FIG. 1, FIG. 1 is a schematic diagram of a memory page Pl write user data Dl. 若用户数据Dl的大小等于4kB而存储页Pl的大小等于16kB,当主机欲通过一存储器控制器将用户数据Dl写入存储页Pl时,用户数据Dl会占据存储页Pl中4kB的空间,其它12kB的记忆空间则必须用来存储其它数据,因而造成此12kB存储器空间的浪费。 If the size of the user data Dl is equal to the size of 4kB page Pl is equal to 16kB memory, when the host user data to be written to the memory page Pl Dl by a memory controller, a user data Dl occupy space in the storage of 4kB page Pl, other 12kB memory space must be used to store other data, resulting in waste of this 12kB memory space. 此外,由于存储器控制器必须一次写入完整的存储页大小(16kB),但仅有4kB的数据须写入,因此程序运作会额外占用存储器控制器的12kB带宽。 In addition, since the memory controller must complete write-once memory page size (16kB), but only 4kB data must be written, so the program will operate additional 12kB occupied bandwidth of the memory controller. 在此情况下,写入效率会大幅降低。 In this case, the writing efficiency will be greatly reduced.

[0004] 在一般非易失性存储器中,数据会在一次写入操作中被写入一存储页,此数据可能为一用户数据(User Data)、对应于用户数据的一映射信息(Mapping Informat1n)、垃圾回收(Garbage Collect1n)数据或平均读写存储区块(Wear-Leveling)数据等。 [0004] In general non-volatile memory, the data is written once in a memory page write operation, this data may be a user data (User Data), corresponding to a user data mapping information (Mapping Informat1n ), garbage collection (garbage Collect1n) or average data write memory block (Wear-leveling) data. 在这些数据中,不同类型的数据会写入不同存储页。 In these data, the different types of data are written to different memory pages. 举例来说,当一区段大小的用户数据欲写入非易失性存储器时,对应于该用户数据的映射信息必须随之而更新,使得系统须分配至少两个存储页来处理该用户数据(一存储页处理用户数据以及另一存储页处理映射信息)。 For example, when the size of a user data sector to be written to the nonvolatile memory, the mapping information corresponding to the user data to be subsequently updated, the system shall be such that at least two memory pages allocated to process the user data (process a memory page storing user data, and another sheet processing mapping information). 在此情况下,除了在存储用户数据的存储页中可能产生浪费之外,系统还须通过另一次写入操作来更新映射信息。 In this case, in addition to wasted memory page may store user data, the system must write operation to update the map information through another. 因此,存储器控制器的写入效率无法有效率地运用。 Therefore, the write efficiency of the memory controller can not efficiently use. 有鉴于此,公知技术实有改进的必要。 In view of this, there is a known technique real need for improvement.

发明内容 SUMMARY

[0005] 因此,本发明的主要目的即在于提供一种可在一次写入一存储页的操作中同时写入数据及映射信息的管理非易失性存储器的方法。 [0005] Accordingly, the main object of the present invention is to provide a method, i.e., data can be written and manage mapping information of the nonvolatile memory while a write-once memory page operations.

[0006] 本发明公开一种管理一非易失性存储器的方法,该非易失性存储器包括有多个存储块,其中每一存储块包括有多个存储页。 [0006] The present invention discloses a management method of a nonvolatile memory, volatile memory comprises a plurality of memory blocks, where each memory block comprises a plurality of memory pages. 该方法包括有将该多个存储页中一存储页分割为多个群组;以及将一数据及对应于该数据的一映射信息写入该多个群组中不同群组。 The method includes storing a plurality of the pages of a memory page is divided into a plurality of groups; and a data corresponding to the data and a mapping information written to the plurality of groups of different groups.

[0007] 本发明还公开一种非易失性存储装置,包括有一非易失性存储器,包括有多个存储块,其中每一存储块包括有多个存储页;以及一存储器控制器,耦接至该非易失性存储器。 [0007] The present invention also discloses a nonvolatile memory device, comprising a non-volatile memory, comprising a plurality of memory blocks, where each memory block comprises a plurality of pages of storage; and a memory controller coupled connected to the non-volatile memory. 该存储器控制器通过执行以下步骤,对该非易失性存储器进行管理:将该多个存储页中一存储页分割为多个群组;以及将一数据及对应于该数据的一映射信息写入该多个群组中不同群组。 The memory controller by performing the following steps, the nonvolatile memory management: a plurality of memory pages in the page memory is divided into a plurality of groups; and a data and a mapping information corresponding to the data write the plurality of groups into different groups.

附图说明 BRIEF DESCRIPTION

[0008] 图1为数据写入一存储页的示意图。 [0008] Figure 1 is a schematic view of data written to a memory page.

[0009] 图2为本发明实施例一非易失性存储装置受控于一主机的示意图。 [0009] FIG. 2 is a schematic embodiment of a nonvolatile memory device is controlled by a host embodiment of the present invention.

[0010] 图3为本发明实施例一存储页的示意图。 [0010] FIG. 3 is a schematic diagram of a embodiment of the present invention, the memory pages.

[0011] 图4为本发明实施例存储页的示意图。 [0011] FIG. 4 is a schematic diagram of the embodiment of the invention the memory pages.

[0012] 图5为本发明实施例超级页的示意图。 [0012] FIG 5 illustrates an embodiment of the present invention, the super-page.

[0013] 图6为本发明实施例一超级页进行备份的示意图。 [0013] FIG. 6 is a schematic diagram of a backup page Super embodiment of the present invention.

[0014] 图7为本发明实施例超级页的示意图。 [0014] FIG. 7 illustrates an embodiment of the present invention, the super-page.

[0015] 其中,附图标记说明如下: [0015] wherein reference numerals as follows:

[0016] 20 非易失性存储装置 [0016] The nonvolatile memory means 20

[0017] 210 存储器控制器 [0017] The memory controller 210

[0018] 220 非易失性存储器 [0018] The nonvolatile memory 220

[0019] 250 主机 [0019] Host 250

[0020] P1、P3、P41、P42、P50 〜P53、P60、存储页 [0020] P1, P3, P41, P42, P50 ~P53, P60, memory pages

[0021] P61、P70 〜P77 [0021] P61, P70 ~P77

[0022] P5A、P5B、P6、P7A、P7B 超级页 [0022] P5A, P5B, P6, P7A, P7B super page

[0023] PLO 〜PL3 层面 [0023] PLO ~PL3 level

[0024] Dl〜D7、UD 用户数据 [0024] Dl~D7, UD User Data

[0025] Ml〜M4、MI 映射信息 [0025] Ml~M4, MI mapping information

[0026] GCl〜GC9、GC 垃圾回收资料 [0026] GCl~GC9, GC garbage collection information

[0027] WLUffL 平均读写存储区块数据 [0027] WLUffL leveling data storage blocks

具体实施方式 Detailed ways

[0028] 请参考图2,图2为本发明实施例一非易失性存储装置20受控于一主机250的示意图。 [0028] Please refer to FIG 2, FIG. 2 embodiment of a nonvolatile memory device embodiment 20 controlled by a host computer 250 in a schematic view of the present disclosure. 如图2所示,非易失性存储装置20包括有一存储器控制器210及一非易失性存储器220。 2, the nonvolatile memory device 20 includes a memory controller 210, and a non-volatile memory 220. 非易失性存储器220由多个存储块(Memory Block)所组成,其中每一存储块由多个存储页(Memory Page)所组成。 The nonvolatile memory 220 by a plurality of memory blocks (Memory Block), where the contents of each memory block a plurality of memory pages (Memory Page) composed. 非易失性存储器220可为一与非门闪速存储器(NAND FlashMemory)或其它类型的非易失性存储器。 The nonvolatile memory 220 may be a flash memory NAND gate (NAND FlashMemory) or other types of nonvolatile memory. 存储器控制器210可用来管理非易失性存储器220。 The memory controller 210 may be used to manage the nonvolatile memory 220. 存储器控制器210可与主机250进行通信,使得主机250可通过存储器控制器210在非易失性存储器220上读取或写入数据。 The memory controller 210 may communicate with the host 250, the host 250 so that data can be read or written by the memory controller 210 on the nonvolatile memory 220. 在一数据写入流程中,主机250通过逻辑存储空间来管理数据,存储器控制器210则用来将逻辑存储空间映射至非易失性存储器220的一物理存储空间,以进行数据管理。 In a data writing process, the host 250 to manage the logical data memory space, the memory controller 210 is used to map logical memory space to a physical memory space in the nonvolatile memory 220, for data management.

[0029] 在公知技术中,当一区段大小的数据(4kB)欲写入一存储页(16kB)时,会浪费存储器控制器210的带宽资源,且非易失性存储器220中多余的存储器空间须用来存储其它数据。 [0029] In the prior art, when a data segment size (4kB) to be written into a memory page (16kB), the memory controller would waste bandwidth resources 210 and 220 redundant memory nonvolatile memory space must be used to store other data. 另一方面,当一存储页的大小等于区段大小时,效率可达到最大。 On the other hand, when the size of a memory segment is equal to the page size, the efficiency can be maximized. 因此,需要根据区段大小,将一存储页切割为多个群组,使得一区段大小的数据可存储在一群组中。 Therefore, according to the segment size, a plurality of memory pages, cut into a group, so that the data size of a segment may be stored in a group. 在此情况下,存储页的物理地址是以群组大小为单位映射至逻辑地址。 In this case, the physical page address stored in units of group size is mapped to the logical address. 举例来说,若一存储页的大小等于16kB而一区段大小等于4kB,较佳地,存储页可分割为4个群组,使得每一区段的数据可存储在一完整的群组。 For example, if the size of a memory page is equal to a segment size equal to 16kB and 4kB, preferably, the memory pages may be divided into four groups, so that the data may be stored in each segment is a complete group. 在存储页中,一物理群组映射至一逻辑群组。 In the memory page, a physical group is mapped to a logical group. 如此一来,每一群组都可用来存储一区段大小的数据,而不浪费任何存储器空间。 As a result, each group can be used to store data for one sector size, without wasting any memory space.

[0030] 为了有效运用存储器空间及存储器控制器210的带宽,每一存储页中的群组都必须存储有效的数据。 [0030] In order to effectively use the bandwidth of the memory controller 210 and memory space, each memory page must be stored in a group of valid data. 在部分实施例中,在每一次写入操作时,欲写入的用户数据长度可能不会等于4个区段大小。 In some embodiments, at each write operation, the length of the user data to be written may not be equal to the size of four segments. 若只有2个区段大小的用户数据欲存储时,存储页中只有2个群组为用户数据所占用。 If only two segment size user data to be stored, only two memory page groups occupied by user data. 在此情况下,另外2个群组可用来存储其它数据,例如映射信息、垃圾回收数据或平均读写存储区块数据等。 In this case, the other two groups can be used to store other data, such as map information, data, or an average garbage storage blocks read data and the like. 因此,写入同一存储页中的数据可包括如用户数据、映射信息、垃圾回收数据及平均读写存储区块数据等不同类型的数据。 Thus, data is written to the same memory page may include different types of data such as user data mapping information, the garbage collection and wear-leveling data storage blocks data. 如此一来,不同类型的数据可在一次写入操作中同时写入一存储页。 Thus, different types of data may be simultaneously written into the first write operation of a memory page.

[0031] 举例来说,请参考图3,图3为本发明实施例一存储页P3的示意图。 [0031] For example, please refer to FIG. 3, FIG. 3 is a schematic diagram of a memory page P3 of the embodiment of the present invention. 如图3所示,3个区段大小的用户数据D2〜D4欲写入存储页P3。 3, the three segments of the user data size to be written in the storage D2~D4 page P3. 由于存储页P3的大小等于16kB而区段长度等于4kB,存储页P3可切割为4个群组。 Since the memory page size P3 is equal to the segment length is equal to 16kB and 4kB, memory page P3 may be cut into four groups. 在存储页P3中,3个群组是用来存储用户数据D2〜D4,而另一群组可用来存储映射信息Ml。 In the memory page P3, the three groups are used to store user data D2~D4, and the other groups are used to store the mapping information Ml. 换句话说,用户数据D2〜D4及映射信息Ml可在一次写入操作中同时写入存储页P3中的不同群组。 In other words, the user data and mapping information D2~D4 Ml may be once written into the memory write operation while the page P3 in different groups.

[0032] 一般来说,映射信息Ml记录了相关于用户数据D2〜D4的逻辑地址映射至存储页P3中相对应群组的物理地址的信息。 [0032] In general, the mapping information recorded Ml D2~D4 logical address information to physical address memory map page group corresponding to P3 in the user data. 映射信息Ml也可包括相关于一物理地址映射至一逻辑地址的信息。 Ml can also include mapping information associated with mapping a physical address to a logical address information. 在部分实施例中,映射信息Ml包括有相关于存储页P3的物理地址映射至用户数据D2〜D4的逻辑地址的信息。 In some embodiments, the mapping information mapped to Ml includes information related to the user data of the logical address physical address stored in D2~D4 page P3. 在部分实施例中,控制器可利用这些映射信息,如逻辑地址映射至物理地址以及物理地址映射至逻辑地址的信息,来进行垃圾回收及/或平均读写存储区块功能。 In some embodiments, the controller may use this mapping information, such as mapping the logical address to the physical address and the physical address information mapped to the logical address, garbage collection is performed and / or read-write memory block average function.

[0033] 值得注意的是,当存储在一特定存储页的映射信息包括有相关于该特定存储页中其它群组所存储数据的映射信息时,此数据与其相对应的映射信息可在一次写入操作中同时更新。 [0033] It is noted that, when the mapping information stored in a particular memory page including mapping information related to other groups of data stored in the particular memory page, the data mapping information corresponding thereto may be write-once the update operation simultaneously. 在此情况下,断电恢复(Power-off Recovery,P0R)的管理会更加容易。 In this case, power restored (Power-off Recovery, P0R) management will be easier. 在一公知非易失性存储器中,数据及其相对应的映射信息必须分别写入非易失性存储器。 In a well-known non-volatile memory, and the data corresponding to the map information to be written to the nonvolatile memory, respectively. 一旦发生断电意外,存储器控制器必须根据数据是否已更新以及相对应的映射信息是否已更新来进行断电恢复。 Once off accident, the memory controller must be updated if the data has power restored, and whether mapping information corresponding to the updated according to. 相较之下,在本发明上述实施例的非易失性存储器中,数据及相对应映射信息可在一次写入操作中同时写入不同群组。 In contrast, in the nonvolatile memory of the present invention, the above-described embodiment, the data and mapping information corresponding to a write operation can be simultaneously written to different groups. 在此情况下,当发生断电意外而必须执行断电恢复时,上述情况可简化为:数据及相对应的映射信息是否已更新;亦即,数据及其相对应的映射信息必然会在同一时间更新,可大幅降低系统执行断电恢复的复杂度。 In this case, when the power recovery must perform an unexpected power failure occurs, the above simplified to: if the data and corresponding map information has been updated; that is, data and corresponding map information is bound in the same time updates, can greatly reduce the complexity of the system power-off recovery.

[0034] 在部分实施例中,可能只会有一区段大小的用户数据欲存储在非易失性存储器中,且相对应的映射信息也必须随之而更新。 [0034] In some embodiments, a segment may only have to be the size of the user data stored in the nonvolatile memory and corresponding to the mapping information will also be updated. 请参考图4,图4为本发明实施例存储页P41及P42的示意图。 Please refer to FIG. 4, FIG. 4 is a schematic embodiment of a memory page P41 and P42 of the present invention. 如图4所示,一区段大小的用户数据D5及其相对应的映射信息M2欲写入存储页P41中。 4, a section the size of the user data D5 and the corresponding mapping information to be written in the memory M2 of the page P41. 由于存储页P41的大小等于16kB而区段长度等于4kB,存储页P41可切割为4个群组。 Since the size of the memory page P41 is equal to segment length equal to 16kB and 4kB, memory page P41 may be cut into four groups. 然而,写入存储页P41的数据仅包括一用户数据D5及其相对应的映射信息M2,这些数据仅占用存储页P41中2个群组,而另外2个群组为多余的。 However, data written in the memory page P41 includes only a user data D5 corresponding to the map information and M2, the data stored in the page P41, with only two groups, while the other two groups is superfluous. 为有效利用存储页P41的存储空间,此2个群组可用来存储垃圾回收数据或平均读写存储区块数据。 Efficient use of storage space to store the page P41, the two groups are used to store data or garbage-leveling data storage blocks. 如图4所示,一区段大小的垃圾回收资料GCl及一区段大小的平均读写存储区块数据WLl与用户数据D5同时写入存储页P41中,映射信息M2也可同时包括对应于垃圾回收数据GCl及平均读写存储区块数据WLl的映射信息。 , The average write data GCl garbage collection size and a section of a section of the data memory block size of the user data D5 WLl simultaneously written in the memory page P41 in FIG. 4, the mapping information may also be included simultaneously M2 corresponding to garbage collection and wear-leveling data storage blocks GCl WLl data mapping information.

[0035] 值得注意的是,欲填入群组的数据大小也可能大于群组大小,因此数据须存储在一个以上的群组,如图4中的存储页P42所示。 [0035] Notably, the size of data to be filled in the group may also be greater than the size of the group, and therefore data to be stored in more than one group, as shown in FIG 4 stored in the page P42. 一区段大小的用户数据D6及其相对应的映射信息M3同时写入存储页P42中。 A segment size of the user data D6 corresponding to the map information and simultaneously written in the memory M3 of the page P42. 存储页P42的大小也等于16kB并分割为4个群组。 Memory page P42 is also equal to the size of 16kB and divided into four groups. 在存储页P42中,2个群组分别用来存储用户数据D6及映射信息M3,而另外2个群组都可用来存储垃圾回收数据GC2。 In the memory page P42, the two groups are used to store user data and mapping information D6 M3, while the other two groups can be used to store data garbage GC2. 若垃圾回收数据GC2的大小大于4kB时(例如等于8kB),垃圾回收数据GC2即可填入存储页P42中此2个群组,使得存储页P42的空间能够有效运用,存储器控制器的效率也可达到最佳化。 If garbage collection when a data size GC2 greater than 4kB (e.g. equal to 8kB), GC2 garbage data to populate memory page P42 in this group 2, so that space for storing the page P42 can be effectively used, the efficiency of the memory controller also can optimally.

[0036] 对一般非易失性存储器而言,垃圾回收是在抹除一存储块以前,将该存储块中的有效数据从一存储页搬移至另一存储块的存储页,平均读写存储区块则是一种数据配置方法使数据的抹除及写入可平均分散在非易失性存储器中,进而延长非易失性存储器的寿命。 [0036] in terms of the general non-volatile memory, garbage collection is a memory block in the erase previously, move the valid data block from a memory pages to the memory block to another memory page, leveling storage it is a data block that the erase and write configuration data may be an average dispersed in non-volatile memory, thereby extending the life of the nonvolatile memory. 在公知非易失性存储器中,相关于垃圾回收及平均读写存储区块运作的资料都必须分别处理。 In known non-volatile memory, information related to garbage collection and wear leveling operation of the memory block must be processed separately. 相较之下,在本发明上述实施例中,由于一存储页切割为多个群组,且每一物理群组都可映射至一逻辑群组,因此垃圾回收数据及平均读写存储区块数据可同时写入一存储页中的不同群组。 In contrast, in the above embodiment of the present invention, since a plurality of memory pages, cut into groups, and each group can be mapped to a physical logical group, thus leveling and garbage data memory blocks data may be written in a memory page different groups simultaneously. 如图4的存储页P41所示,当2个群组分别填入用户数据D5及映射信息M2以后,另外2个群组可用来存储垃圾回收数据GCl及平均读写存储区块数据WL1,使得存储器控制器可进行垃圾回收及平均读写存储区块的运作,而此垃圾回收及平均读写存储区块是原先系统预设欲在一段时间以后进行的工作。 FIG memory pages of P41 4, when two groups are populated user data and mapping information after M2 D5, the other two groups are used to store data GCl garbage collection and wear-leveling data storage blocks WL1, such that the memory controller can operate garbage collection and an average read and write memory block, and this garbage collection and wear-leveling system default memory block is the original work to be done at a later time. 如此一来,可提升系统执行垃圾回收及平均读写存储区块的效率。 In this way, can enhance the efficiency of garbage collection and wear-leveling block storage systems.

[0037] 值得注意的是,上述实施例也可应用于多层面写入(Mult1-Plane Programming)的情况。 [0037] It is noted that the above embodiments are also applicable to the case of writing multidimensional (Mult1-Plane Programming) a. 在部分实施例中,在不同层面(Plane)的不同存储块中数个存储页可结合为一超级页(Super Page)。 In some embodiments, at different levels (Plane) different number of memory blocks may be combined into a memory page superpage (Super Page). 请参考图5,图5为本发明实施例超级页P5A及P5B的示意图。 Please refer to FIG. 5, FIG. 5 is a schematic diagram of the super-page P5A and P5B embodiment of the invention. 如图5所示,超级页P5A由存储页P50及P51所组成,其中存储页P50位于一层面PLO而存储页P51位于一层面PL1。 5, the super P5A page P50 and P51 are stored in the page, where the page P50 is stored is located at a level PLO page P51 is stored is located at a level PL1. 每一存储页的大小等于8kB,而主机用来管理非易失性存储器的区段单位等于4kB,因此每一存储页P50及P51可分割为2个群组。 Each memory page size equal to 8kB, and the host to manage the nonvolatile memory section unit is equal 4kB, so that each memory page P50 and P51 may be divided into two groups. 在一次写入操作中,数据可同时写入超级页P5中的存储页P50及P51,因此4个区段大小的数据即可完整写入超级页,其中,写入的数据可为映射信息M1、用户数据UD、垃圾回收数据GC或平均读写存储区块数据WL,如图5的超级页P5A所示。 In the write-once operation, data can be simultaneously written memory page P50 and P51 Super page (P5), so the size of the data segments 4 to super-page writes complete, wherein the data mapping information may be written to M1 , the UD user data, or the garbage collection GC-leveling data storage blocks data WL, as shown in Figure 5 of the super-page P5A.

[0038] 同样地,当数据大小大于群组大小时,须使用一个以上的群组来存储数据,如超级页P5B所示。 [0038] Similarly, when the data size is greater than the size of the group, the group must be used to store one or more data, as shown in the super page P5B. 超级页P5B由存储页P52及P53所组成,其中存储页P52位于层面PLO而存储页P53位于层面PLl。 Super page P5B by the memory page P52 and P53, where the contents of the memory pages located P52 level PLO and memory page P53 located level PLl. 每一存储页的大小等于8kB,而主机用来管理非易失性存储器的区段单位等于4kB,因此每一存储页P52及P53可分割为2个群组。 Each memory page size equal to 8kB, and the host to manage the nonvolatile memory section unit is equal 4kB, so that each memory page P52 and P53 may be divided into two groups. 映射信息及用户数据分别存储在存储页P52的一群组。 Mapping information and user data are stored in a group of memory page P52. 垃圾回收数据GC3的大小等于8kB,因此必须占用超级页P5B中2个群组。 GC3 garbage collection data size is equal to 8kB, it is necessary to take up the super page P5B in two groups. 如图5所示,存储页P53的2个群组都用来存储垃圾回收数据GC3。 5, two groups page P53 are stored in the garbage collector to store data GC3.

[0039] 在应用于超级页的实施例中,超级页中不同存储页之间可进行备份。 [0039] In the embodiment applied to the super-page, the page can be backed super between different memory pages. 请参考图6,图6为本发明实施例一超级页P6进行备份的示意图。 Please refer to FIG. 6, FIG. 6 is a schematic diagram of a super-backup page P6 embodiment of the present invention. 如图6所示,超级页P6由存储页P60及P61所组成,其中存储页P60位于层面PLO而存储页P61位于层面PLl。 6, a super-page P6 P60 and P61 are stored in the page composition, which is located in memory page P60 P61 level PLO stored page level located PLl. 每一存储页的大小等于8kB,而主机用来管理非易失性存储器的区段单位等于4kB,因此每一存储页P60及P61可分割为2个群组。 Each memory page size equal to 8kB, and the host to manage the nonvolatile memory section unit is equal 4kB, so that each memory page P60 and P61 may be divided into two groups. 当用户数据D7欲写入超级页P6时,用户数据D7可同时写入存储页P60及P61中,以进行备份。 When the Supervisor P6 user data of page D7 to be written, the user data can be simultaneously written D7 P60 and P61, the memory pages to be backed up. 对应于用户数据D7的映射信息M4也可同时在存储页P60及P61内更新。 M4 D7 mapping information can also be updated at the same time corresponding to the user data in the memory page P60 and P61. 在此情况下,超级页P6中所有群组都能够有效运用,而不浪费存储页中任何空间,也不浪费存储器控制器的带宽资源。 In this case, the super page P6 in all groups are able to use effectively, without wasting any space to store page, do not waste bandwidth memory controller.

[0040] 在公知非易失性存储系统中,当用户数据欲写入非易失性存储器时,通常会在主机的寄存器中保留用户数据以进行备份,直到写入的动作完成为止。 [0040] When user data is to be written to the nonvolatile memory, usually preserved in the known nonvolatile memory in the host system user data registers for backup until the writing operation is completed. 在此情况下,在数据写入期间内,数据必须持续存储在寄存器中。 In this case, during data writing, data must be continuously stored in the register. 相较之下,若用户数据在一次写入操作中同时写入2个存储页,其中一存储页即可视为备份。 In contrast, if the user data is written to two memory pages in a write operation, wherein a memory page may be regarded as a backup. 在此情况下,主机的寄存器在主机对存储器控制器下命令完成以后即可释放出来,可大幅降低所需的寄存器空间大小。 In this case, the master registers to be released after the completion of the host command to the memory controller, can significantly reduce the required size of the register space.

[0041] 请参考图7,图7为本发明实施例超级页P7A及P7B的示意图。 [0041] Please refer to FIG. 7, FIG. 7 is a schematic diagram of the super-page P7A and P7B embodiment of the present invention. 如图7所示,超级页P7A及P7B具有4层面结构。 7, the super-page P7A and P7B 4 having the layer structure. 超级页P7A包括有4个存储页P70〜P73,分别位于层面PLO〜PL3,超级页P7B包括有4个存储页P74〜P77,分别位于层面PLO〜PL3。 Super P7A page includes four memory pages P70~P73, are located level PLO~PL3, super P7B page includes four memory pages P74~P77, it is located level PLO~PL3. 每一存储页的大小都等于8kB,而主机用来管理非易失性存储器的区段单位等于4kB,因此每一存储页P70〜P77可分割为2个群组,其中每一群组大小等于4kB。 Each memory page size equal to 8kB, and the host to manage the nonvolatile memory unit is equal to 4kB sector, so that each memory page P70~P77 may be divided into two groups, with each group size is equal to 4kB. 即使只有一用户数据欲存储在超级页P7A或P7B而使得超级页P7A或P7B中只有一群组存储用户数据,其它群组都可用来填补垃圾回收数据以及相对应的映射信息。 Even if only a super-user data to be stored in the page or P7A P7B such that only a group of user data is stored in a super page P7A or P7B, other groups can be used to fill the garbage data and the mapping information corresponding to. 在超级页P7A中,垃圾回收资料GC4的大小等于8kB并占用存储页P72的2个群组,垃圾回收资料GC5的大小等于8kB并占用存储页P73的2个群组。 In the super-page P7A, the garbage collection data is equal to the size of GC4 8kB memory page P72 and occupy the two groups, garbage collection data is equal to the size of GC5 8kB and occupy two groups of memory page P73. 由于映射信息同时对应于用户数据以及垃圾回收数据GC4及GC5,因此,映射信息也可能需要使用更多群组来存储。 Since the mapping information corresponding to the user data, and at the same time garbage data GC4 and GC5 are, therefore, it requires the use of mapping information may also be stored more groups. 在此情况下,存储页P70的2个群组及存储页P71的I群组可用来存储映射信息。 In this case, two memory pages of the group P70 and P71 stored in the I group of pages used to store the mapping information. 在超级页P7B中,每一垃圾回收资料GC6〜GC9的大小等于4kB,垃圾回收数据GC6〜GC9并分别填入存储页P76或P77中一群组。 In the super-page P7B, the size of each garbage collection information GC6~GC9 equal to 4kB, garbage collection and data GC6~GC9 were filled memory page P76 or P77 in a group. 由于映射信息同时对应于用户数据以及垃圾回收数据GC6〜GC9,因此映射信息也可能需要使用更多群组来存储。 Since the mapping information corresponding to the user data, and at the same time garbage collection data GC6~GC9, thus mapping information you may need to use more storage groups. 在此情况下,存储页P74的2个群组及存储页P75的I群组可用来存储映射信息。 In this case, two groups memory page P75 and P74 of memory pages used to store the I group mapping information.

[0042] 值得注意的是,本发明可根据主机用来管理非易失性存储器的区段大小,将一存储页分割为多个群组,并在不同群组中写入不同类型的数据。 [0042] It is noted that, according to the present invention may be used to manage the segment size of the host non-volatile memory, the memory page is divided into a plurality of groups, and written to different types of data in different groups. 本领域的技术人员当可据以修饰或变化,而不限于此。 Those skilled in the art to modify or when the data can be changed without being limited thereto. 举例来说,在一次写入操作中,主机可根据系统需求,写入任何类型的数据,其包括但不限于用户数据、垃圾回收数据、平均读写存储区块数据及对应于上述每一种数据的映射信息。 For example, in a write operation, the host according to the system requirements, write any type of data, including, but not limited to, user data, garbage data, leveling data of the memory block and corresponding to each of the above mapping information data. 除此之外,数据在不同群组中的配置方式也可根据系统需求而定,例如断电恢复或备份的需求。 In addition, the data configuration in the different groups may also be determined according to system requirements, such as requirements or backup power recovery.

[0043] 在公知技术中,当一存储页中一区段大小的数据欲更新时,此区段的更新数据必须写入一新的存储页,而在新的存储页中其它多余的记忆空间则必须填补原存储页中其它数据,因而造成存储器空间的浪费以及写入效率的低落。 [0043] In the prior art, when a section of a memory page size of data to be updated, to update data in a segment to be written to the new memory page, and the page in the new storage space in the memory of other unwanted it must fill the pages of the original stored in other data, resulting in a waste of memory space and writing efficiency is low. 此外,在每一次写入操作中,一存储页只能写入一种类型的数据。 Further, each write operation, a memory page can be written only one type of data. 相较之下,本发明可将存储页切割为多个群组,而每一逻辑群组可映射至一物理群组。 In contrast, the present invention may be cut into a plurality of groups of memory pages, and each logical group may be mapped to a physical group. 不同类型的数据可存储在同一存储页的不同群组中,可提升存储器控制器的带宽使用效率以及存储器的空间使用效率。 Different types of data may be stored in different groups in the same memory page, can improve the efficiency of bandwidth usage and memory space of the memory controller efficiency. 除此之外,用户数据备份及断电恢复管理的效率也可同时获得提升。 In addition, user data backup and recovery management efficiency power outage also be promoted at the same time.

[0044] 以上所述仅为本发明的优选实施例而已,并不用于限制本发明,对于本领域的技术人员来说,本发明可以有各种更改和变化。 [0044] The above description is only preferred embodiments of the present invention, it is not intended to limit the invention to those skilled in the art, the present invention may have various changes and variations. 凡在本发明的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本发明的保护范围之内。 Any modification within the spirit and principle of the present invention, made, equivalent substitutions, improvements, etc., should be included within the scope of the present invention.

Claims (20)

1.一种管理一非易失性存储器的方法,该非易失性存储器包括有多个存储块,其中每一存储块包括有多个存储页,该方法包括有: 将该多个存储页中一存储页分割为多个群组;以及将一数据及对应于该数据的一映射信息写入该多个群组中不同群组。 1. A method of managing method of a nonvolatile memory, the nonvolatile memory includes a plurality of memory blocks, where each memory block comprises a plurality of memory pages, the method comprising: storing the plurality of pages in a memory page is divided into a plurality of groups; and a data corresponding to the data and a mapping information written to the plurality of groups of different groups.
2.如权利要求1所述的方法,其特征在于,该方法还包括有: 以一群组的大小为单位来映射一逻辑地址至一物理地址。 2. The method according to claim 1, characterized in that, the method further comprising: a size of a group as a unit to map a logical address to a physical address.
3.如权利要求1所述的方法,其特征在于,将该数据及对应于该数据的该映射信息写入该多个群组中不同群组的步骤在一次写入该存储页的操作中进行。 3. The method as recited in claim 1, wherein the data corresponding to the data and the writing operation of the mapping information in the plurality of groups of steps in different groups of the write-once memory pages get on.
4.如权利要求1所述的方法,其特征在于,该数据包括一用户数据、一垃圾回收数据或一平均读写存储区块数据。 4. The method according to claim 1, wherein the data comprises a user data, a garbage collection or data stored in a leveling block data.
5.如权利要求1所述的方法,其特征在于,该映射信息包括相关于该数据的一逻辑地址映射至该非易失性存储器的一物理地址的信息。 5. The method according to claim 1, wherein the information includes information related to a logical address of the data is mapped to the non-volatile memory of a physical address of the mapping.
6.如权利要求1所述的方法,其特征在于,该映射信息包括相关于该数据的一物理地址映射至一逻辑地址的信息。 6. The method according to claim 1, wherein the mapping information includes a physical address associated with the data information is mapped to a logical address.
7.如权利要求1所述的方法,其特征在于,该方法还包括有: 结合不同存储块中的存储页,以产生一超级页。 7. The method according to claim 1, characterized in that, the method further comprising: binding different memory pages of the memory blocks, to produce a super page.
8.如权利要求7所述的方法,其特征在于,该方法还包括有: 将该数据分别写入该超级页中不同存储页。 8. The method according to claim 7, characterized in that, the method further comprising: the data pages are written to the super different memory pages.
9.如权利要求8所述的方法,其特征在于,该数据包括一用户数据。 9. The method according to claim 8, wherein the data comprises a user data.
10.如权利要求1所述的方法,其特征在于,该数据存储在该存储页的该多个群组中至少一群组。 10. The method according to claim 1, characterized in that at least one group of the data stored in the plurality of groups of the memory page.
11.一种非易失性存储装置,包括有: 一非易失性存储器,包括有多个存储块,其中每一存储块包括有多个存储页;以及一存储器控制器,耦接至该非易失性存储器,通过执行以下步骤,对该非易失性存储器进行管理: 将该多个存储页中一存储页分割为多个群组;以及将一数据及对应于该数据的一映射信息写入该多个群组中不同群组。 A nonvolatile memory device, comprising: a nonvolatile memory including a plurality of memory blocks, where each memory block comprises a plurality of pages of storage; and a memory controller, coupled to the non-volatile memory, by performing the following steps, the nonvolatile memory management: a plurality of memory pages in the page memory is divided into a plurality of groups; and a data and data corresponding to a map of the information written in the plurality of groups of different groups.
12.如权利要求11所述的非易失性存储装置,其特征在于,该存储器控制器还执行以下步骤,以对该非易失性存储器进行管理: 以一群组的大小为单位来映射一逻辑地址至一物理地址。 12. The nonvolatile memory device according to claim 11, wherein the memory controller further performs the following steps to manage the nonvolatile memory: a size of a group is mapped in units of a logical address to a physical address.
13.如权利要求11所述的非易失性存储装置,其特征在于,该数据及对应于该数据的该映射信息在一次写入该存储页的操作中写入该多个群组中不同群组。 13. A nonvolatile memory device according to claim 11, wherein the data corresponding to the data and the mapping information in the write-once plurality of different groups of the write operation of the memory pages of group.
14.如权利要求11所述的非易失性存储装置,其特征在于,该数据包括一用户数据、一垃圾回收数据或一平均读写存储区块数据。 14. The nonvolatile memory device according to claim 11, wherein the data comprises a user data, a garbage collection or data stored in a leveling block data.
15.如权利要求11所述的非易失性存储装置,其特征在于,该映射信息包括相关于该数据的一逻辑地址映射至该非易失性存储器的一物理地址的信息。 15. The nonvolatile memory device according to claim 11, wherein the information includes information of a physical address associated with a logical address of the data is mapped to the non-volatile memory of the map.
16.如权利要求11所述的非易失性存储装置,其特征在于,该映射信息包括相关于该数据的一物理地址映射至一逻辑地址的信息。 16. The nonvolatile memory device according to claim 11, wherein the mapping information includes information on mapping a logical address to a physical address in the data.
17.如权利要求11所述的非易失性存储装置,其特征在于,该存储器控制器还执行以下步骤,以对该非易失性存储器进行管理: 结合不同存储块中的存储页,以产生一超级页。 17. The nonvolatile memory device according to claim 11, wherein the memory controller further performs the following steps to manage the nonvolatile memory: binding different memory blocks of memory pages to generating a super page.
18.如权利要求17所述的非易失性存储装置,其特征在于,该存储器控制器还执行以下步骤,以对该非易失性存储器进行管理: 将该数据分别写入该超级页中不同存储页。 18. The nonvolatile memory device according to claim 17, wherein the memory controller further performs the following steps to manage the nonvolatile memory: the data are written to the super-page different storage page.
19.如权利要求18所述的非易失性存储装置,其特征在于,该数据包括一用户数据。 19. The nonvolatile memory device according to claim 18, wherein the data comprises a user data.
20.如权利要求11所述的非易失性存储装置,其特征在于,该数据存储在该存储页的该多个群组中至少一群组。 20. The nonvolatile memory device according to claim 11, characterized in that at least one group of the data stored in the plurality of groups of the memory page.
CN 201310547079 2013-05-02 2013-11-06 Method of managing non-volatile memory and non-volatile storage device thereof CN104133774A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US201361818884P true 2013-05-02 2013-05-02
US13/953,764 US20140328127A1 (en) 2013-05-02 2013-07-30 Method of Managing Non-Volatile Memory and Non-Volatile Storage Device Using the Same

Publications (1)

Publication Number Publication Date
CN104133774A true CN104133774A (en) 2014-11-05

Family

ID=51806457

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201310547079 CN104133774A (en) 2013-05-02 2013-11-06 Method of managing non-volatile memory and non-volatile storage device thereof

Country Status (1)

Country Link
CN (1) CN104133774A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106897026A (en) * 2016-10-07 2017-06-27 威盛电子股份有限公司 Nonvolatile memory device and its address classification method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080270680A1 (en) * 2005-11-17 2008-10-30 Chee Keng Chang Controller for Non-Volatile Memories and Methods of Operating the Memory Controller
US20090168525A1 (en) * 2007-12-27 2009-07-02 Pliant Technology, Inc. Flash memory controller having reduced pinout
CN101739344A (en) * 2008-11-17 2010-06-16 超捷公司 Memory controller for controlling the wear in a non-volatile memory device and a method of operation therefor
US20110145483A1 (en) * 2009-12-16 2011-06-16 Samsung Electronics Co., Ltd. Semiconductor memory device and method of processing data for erase operation of semiconductor memory device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080270680A1 (en) * 2005-11-17 2008-10-30 Chee Keng Chang Controller for Non-Volatile Memories and Methods of Operating the Memory Controller
US20090168525A1 (en) * 2007-12-27 2009-07-02 Pliant Technology, Inc. Flash memory controller having reduced pinout
CN101739344A (en) * 2008-11-17 2010-06-16 超捷公司 Memory controller for controlling the wear in a non-volatile memory device and a method of operation therefor
US20110145483A1 (en) * 2009-12-16 2011-06-16 Samsung Electronics Co., Ltd. Semiconductor memory device and method of processing data for erase operation of semiconductor memory device

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106897026A (en) * 2016-10-07 2017-06-27 威盛电子股份有限公司 Nonvolatile memory device and its address classification method

Similar Documents

Publication Publication Date Title
USRE44052E1 (en) Flash memory management method
JP5728672B2 (en) Hybrid memory management
EP2920697B1 (en) Methods, data storage devices and systems for fragmented firmware table rebuild in a solid state drive
US8037232B2 (en) Data protection method for power failure and controller using the same
JP5530012B2 (en) Method and system for storage address remapping for storage
US8122193B2 (en) Storage device and user device including the same
US8321652B2 (en) Process and method for logical-to-physical address mapping using a volatile memory device in solid state disks
US7426623B2 (en) System and method for configuring flash memory partitions as super-units
JP5026213B2 (en) Storage device and a deduplication method
JP4931810B2 (en) fat analysis for the management of optimized sequential cluster
US9104315B2 (en) Systems and methods for a mass data storage system having a file-based interface to a host and a non-file-based interface to secondary storage
US8566504B2 (en) Dynamic metablocks
CN101147133B (en) Method for reprogramming non-volatile memory system and memories
US9134918B2 (en) Physical compression of data with flat or systematic pattern
CN100487672C (en) Method and apparatus for splitting a logical block
KR101324688B1 (en) Memory system having persistent garbage collection
US9177638B2 (en) Methods and devices for avoiding lower page corruption in data storage devices
US7840617B2 (en) Host device and memory system
US8683173B2 (en) Logical address offset in response to detecting a memory formatting operation
US7444462B2 (en) Methods for phased garbage collection using phased garbage collection block or scratch pad block as a buffer
KR101288408B1 (en) A method and system for facilitating fast wake-up of a flash memory system
KR101464338B1 (en) Data storage device, memory system, and computing system using nonvolatile memory device
US7254668B1 (en) Method and apparatus for grouping pages within a block
US7395384B2 (en) Method and apparatus for maintaining data on non-volatile memory systems
US8725946B2 (en) Mass storage system and method of using hard disk, solid-state media, PCIe edge connector, and raid controller

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
WD01