(19)
(11)EP 1 576 478 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
28.04.2010 Bulletin 2010/17

(21)Application number: 03749499.4

(22)Date of filing:  10.09.2003
(51)International Patent Classification (IPC): 
G06F 12/00(2006.01)
(86)International application number:
PCT/US2003/028047
(87)International publication number:
WO 2004/040453 (13.05.2004 Gazette  2004/20)

(54)

METHOD AND APPARATUS FOR GROUPING PAGES WITHIN A BLOCK

VERFAHREN UND VORRICHTUNG ZUR SEITENGRUPPIERUNG IN EINEM BLOCK

PROCEDE ET APPAREIL POUR GROUPER DES PAGES DANS UN BLOC


(84)Designated Contracting States:
AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LI LU MC NL PT RO SE SI SK TR

(30)Priority: 28.10.2002 US 281855

(43)Date of publication of application:
21.09.2005 Bulletin 2005/38

(73)Proprietor: SanDisk Corporation
Milpitas, CA 95035 (US)

(72)Inventors:
  • CHANG, Robert, C.
    Danville, CA 94506 (US)
  • QAWAMI, Bahman
    San Jose, CA 95138 (US)
  • SABET-SHARGHI, Farshid
    Los Altos Hills, California 94024 (US)

(74)Representative: Hitchcock, Esmond Antony et al
Marks & Clerk LLP 90 Long Acre
London WC2E 9RA
London WC2E 9RA (GB)


(56)References cited: : 
US-A- 5 404 485
US-A1- 2002 099 904
US-A- 5 937 425
  
  • MEI-LING CHIANG ET AL: "Managing flash memory in personal communication devices" CONSUMER ELECTRONICS, 1997. ISCE '97., PROCEEDINGS OF 1997 IEEE INTERNATIONAL SYMPOSIUM ON SINGAPORE 2-4 DEC. 1997, NEW YORK, NY, USA,IEEE, US, 2 December 1997 (1997-12-02), pages 177-182, XP010268685 ISBN: 0-7803-4371-9
  • KAWAGUCHI A ET AL: "FLASH-MEMORY BASED FILE SYSTEM" PROCEEDINGS OF THE USENIX ANNUAL TECHNICAL CONFERENCE, 1995, pages 155-164, XP000920702
  
Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


Description

BACKGROUND OF THE INVENTION


1. Field of Invention



[0001] The present invention relates generally to mass digital data storage systems. More particularly, the present invention relates to systems and methods for efficiently writing to a block of flash memory.

2. Description of the Related Art



[0002] The use of non-volatile memory systems such as flash memory storage systems is increasing due to the compact physical size of such memory systems, and the ability for non-volatile memory to be repetitively reprogrammed. The compact physical size of flash memory storage systems facilitates the use of such storage systems in devices which are becoming increasingly prevalent. Devices which use flash memory storage systems include, but are not limited to, digital cameras, digital camcorders, digital music players, handheld personal computers, and global positioning devices. The ability to repetitively reprogram non-volatile memory included in flash memory storage systems enables flash memory storage systems to be used and reused.

[0003] In general, flash memory storage systems may include flash memory cards and flash memory chip sets. Flash memory chip sets generally include flash memory components and a controller components. Typically, a flash memory chip set may be arranged to be assembled into an embedded system. The manufacturers of such assemblies or host systems typically acquire flash memory in component-form, as well as other components, then assemble the flash memory and the other components into a host system.

[0004] As shown in Fig.la, within a file system, memory 10 is effectively divided into a system or directory area 12 and a data area 14. System area 12 generally includes root directories and file allocation tables (FATs), while data files are typically included in data area 14. A file system may write data in sectors, e. g., one page at a time, into system area 12, while writing data in clusters, e. g., multiple pages at a time, into data area 14. The size of each cluster, which is generally the minimum size of a file that is associated with data area 14, may vary. For example, in an overall system which is running either a Windows or a DOS system, a cluster may be sized to include approximately four to approximately 256 pages.

[0005] Any updates associated with a logical block, or a block that is associated with a file system, are effectively propagated to the flash media where a physical block is mapped to the logical block. Fig. 1b is a diagrammatic representation of a mapping between a logical block and a physical block of a flash media. A logical block 52 includes a plurality of pages which are each mapped to pages of a physical block 54 such that when a page associated with logical block 52 is updated, the update is written into physical block 54.

[0006] When a page, e. g., a logical page, is updated, a physical block with which the page is associated is written to. Conventionally, a block may be written to ina"fixed"manner or in a"random"manner. When a block is written to in a fixed manner, a page is written into a block at a corresponding page offset. As shown in Fig. 2a, when a logical block page 214 is updated, contents associated with the logical block page 214 are generally written into a corresponding page 218 within a physical block 210. For example, when logical block page 214 corresponds to physical block page 218h, and physical block page 218h is available, the contents of logical block page 214 may be written into physical block page 218h. When physical block page 218h is available, the indication is that there have effectively been no pages 218 written in block 210 that have a higher page number than physical block page 218h.

[0007] In the event that there have been pages 218 written in block 210 that have a higher page number than a logical block page with contents which are to be written, then when block 210 is arranged to be written to in a fixed manner, the contents of the logical block page may not be written into block 210. To write a lower page number after the higher page has been written into the flash memory isan"out-of-sequence"write operation. As will be understood by those skilled in the art, some flash media do not allow for this type of operation. Therefore, the contents of the logical block page as well as contents of pages 218 are written into a new block. Fig. 2b is a diagrammatic representation of an old block, e. g., block210 of Fig. 2a, and a new block which is arranged to accommodate the contents of the old block. When contents of a logical block page 234 are to be written into a physical block in a fixed manner, it is determinedwhether block 210, i. e. , a current block, may be written into. If logical page 234 has a lower page number than is available in block 210, then the contents of logical page 234 may not be written into block 210. In other words, if a physical page 218 with a corresponding page offset to logical block 234 is not available, e. g., has already been written to, then a new physical page 230 generally must be obtained to enable the contents of logical page 234 to be written.

[0008] Since logical block 234 corresponded to page 218c of physical block 210, when block 230 is obtained, the contents of pages 218 are generally copied into corresponding pages 238 of physical block 230, except for the contents of page 218c. Instead of copying the contents of page 218c into page 238c, the contents of logical block 234 are copied into page 238c. Hence, after a copy or write operation, page238a, page238b, and pages 238d-238g contain the contents of page 218a, page 218b, and pages 218d-g, respectively, while page 238c contains the contents associated with logical page 234. Once block 230 has been written to, then block 210 may be erased.

[0009] Writing to a block in a fixed manner is generally efficient when a page with a corresponding page offset is available within the block for a page that is being written. However, when a page with a corresponding page offset is not available to be written into, then even if there are available pages within the block, a new block generally must be obtained to enable the write operation to be completed. Obtaining a new block consumes a significant amount of computational overhead, copy operations which copy contents into the new block and erase operations which erase an old block typically must be run. As will appreciated by those skilled in the art, such operations, particularly write and erase operations, are relatively expensive.

[0010] In lieu of a fixed manner or writing to a block, a random manner may instead be implemented to write data to a block. Contents associated with substantially any logical page may be written into the next available page space in a block when blocks are written to in a random manner. With reference to Fig. 3, a random manner of writing data to a block will be described. When contents of a logical block page 314 are to be written into a physical block 310 and there are available pages 318 within block 310, then the contents of logical block page 314 may be written into the next sequential available page 318. For example, the contents of logical page 314a may be written into page 318h, while the contents of logical page 314b may be written into page 318i.

[0011] It should be appreciated that although logical page 314b may correspond to page 318c, updated contents of logical page 314b may be written into page 318i if page 318i is the next available page in block 310. When logical page 314 is essentially to be read, then it is generally necessary to identify page 318i within block 310 as containing the latest contents associated with logical page 314b. As shown, when page 318i is the last page in block 310, identifying page 318i as containing contents associated with logical page 314b may require reading all of block310 and, hence, may require a relatively high number of read operations. Reading through substantially all of block310 in order to access contents associated with logical page314b may be relatively time-consuming. Although there are techniques which use overall system memory or flash memory for caching information relating to pages 318 and contents of pages 318 such that the number of read operations needed to locate and access contents associated with logical pages such as logical pages 314 may be reduced, such techniques increase memory requirements of the overall system or, more specifically, software executing within the overall system.

[0012] US-A-5937425 describes a method for organizing a flash memory in which the size of the memory portion for reading or writing data, such as a block, differs from the size of the smallest portion for erasing, such as a unit. In order to read a particular logical block or page in the memory, it is necessary to find the last non-free physical block at the same block offset as the desired logical block to be read. US 2002/0099904 also describes a digital data storage system.

[0013] Therefore, what are desired are a hybrid method and an apparatus for efficiently enabling pages associated with physical blocks to be written and to be accessed. That is, what is needed is a hybrid system which utilizes both fixed and random processes to enable pages within a block to be efficiently written to and easily accessed, substantially without requiring relatively frequent block erasures or a relatively large number of read operations.

SUMMARY OF THE INVENTION



[0014] The present invention relates to a system and a method for efficiently enabling pages within a block to be accessed. According to a first aspect of the present invention there is provided A method for writing data into a first block associated with a non-volatile memory, the first block including a plurality of pages, the plurality of physical pages, the plurality of physical pages being grouped into a plurality of physical groups, wherein each physical group of the plurality of physical groups includes two or more physical pages, the method comprising:

identifying a first set of data associated with a first logical page, the first logical page being one of a plurality of logical pages included in a first logical group of a logical block;

determining whether a first physical group of the plurality of physical groups is associated with the first logical group;

writing the first set of data into the first physical page responsive to determining that the first physical page is available to accommodate the first set of data;

determining whether a second physical group in the first physical block is available responsive to determining that the first physical group is not associated with the first logical group;

associating the second physical group with the first logical group responsive to determining that the second physical group is available; and

then writing the first set of data into a second physical page associated with the second physical group.



[0015] Grouping pages within blocks such that a group of pages within a physical block may be obtained in lieu of new physical block when a page is to be written enables the overall efficiency associated with writing pages to be improved. In other words, the likelihood that a new block will have to be obtained in order to enable data to be written may be reduced because of the possibility that a group within the new block may be available. As such, a write process may occur more efficiently, at least in part because the occurrences of erase processes, e.g., an erase processes which are used to erase old blocks once new blocks are obtained, may be significantly reduced. In addition, a read process to read a set of data stored in a particular page may occur more efficiently, as groups may be perused in an attempt to identify the particular page in lieu of reading an entire block to locate the particular page. For example, if a group of pages is determined to not include the particular page, the pages of the group need not be individually read, i. e. , substantially only the pages of the group which includes the particular page may need to be read. By applying a group concept, a read operation is faster than it would be when read operations occur in an effectively random manner, because substantially only the pages of the identical group need to be checked. A write operation which is associated with a group concept is faster than a conventional fixed process, because there are more locations for the page to be written in a single physical block. Thus, there is a reduction in the need to allocate a new block and to erase the old block.

[0016] According to a second aspect of the present invention, A system for writing data into a first physical block associated with a non-volatile memory, the first physical block including a plurality of physical pages, the plurality of physical pages being grouped into a plurality of physical groups, wherein each physical group of the plurality of physical groups includes two or more physical pages, the system comprising:

means for identifying a first set of data associated with a first logical page, the first logical page being one of a plurality of logical pages included in a first logical group of a logical block;

means for determining whether a first physical group of the plurality of physical groups is associated with the first logical group; means for determining whether a first physical page included in the first physical group is available to accommodate the first set of data responsive to determining that the first physical group is associated with the first logical group; and

means for writing the first set of data into the first physical page responsive to determining that the first physical page is available responsive to determining that the first physical group is not associated with the first logical group;

means for associating the second physical group with the first logical group responsive to determining that the second physical group is available; and

means for then writing the first set of data into a second physical page associated with the second physical group.



[0017] These and other advantages of the present invention will become apparent upon reading the following detailed descriptions and studying the various figures of the drawings.

BRIEF DESCRIPTION OF THE DRAWINGS



[0018] The invention may best be understood by reference to the following description of exemplary embodiments taken in conjunction with the accompanying drawings in which:

Fig. 1a is a diagrammatic representation of a memory as effectively divided by a file system.

Fig. 1b is a diagrammatic representation of a mapping between a logical block and a physical block.

Fig. 2a is a diagrammatic representation of a physical block which is arranged to accommodate contents associated with when a logical block page in a fixed manner.

Fig. 2b is a diagrammatic representation of an old physical block, e. g., block210 of Fig. 2a, and a new physical block which is arranged to accommodate the contents of the old block.

Fig. 3 is a diagrammatic representation of a physical block into which contents associated with a logical block page are to be written in a random manner.

Fig. 4a is a diagrammatic representation of a general host system which includes a non- volatile memory.

Fig. 4b is a diagrammatic representation a memory device, e. g., memory device 120 of Fig. 4a.

Fig. 4c is a diagrammatic representation of a host system which includes an embedded non-volatile memory.

Fig. 5 is a diagrammatic representation of a file system with logical blocks and a media with physical blocks in accordance with an embodiment of the present invention.

Fig. 6 is a diagrammatic representation of a block with pages which are arranged into groups in accordance with an embodiment of the present invention.

Fig. 7 is a diagrammatic representation of a page in accordance with an embodiment of the present invention.

Fig. 8a is a diagrammatic representation of a logical block and a physical block which are each divided into groups in accordance with an embodiment of the present invention.

Fig. 8b is a diagrammatic representation of a logical block and a physical block, i.e., logical block 802 and physical block 806 of Fig. 8a, which indicates how groups within the physical block may be written to in accordance with an embodiment of the present invention.

Fig. 8c is a diagrammatic representation of a logical block and a physical block, i.e., logical block 802 and physical block 806 of Fig. 8a, which indicates how groups within the blocks may be associated in accordance with an embodiment of the present invention.

Fig. 8d is a diagrammatic representation of the associations between a logical block, i.e., logical block 802 of Figs. 8a-c, and a new physical block after a merge operation in accordance with an embodiment of the present invention.

Fig. 9 is a diagrammatic representation of a current working block information table in accordance with an embodiment of the present invention.

Figs. 10a and 10b are a process flow diagram which illustrates the steps associated with one method of writing contents associated with a logical page which is part of a logical group into a physical page which is part of a physical group in accordance with an embodiment of the present invention.

Fig. 11 is a diagrammatic block diagram representation of a system architecture in accordance with an embodiment of the present invention.


DETAILED DESCRIPTION OF THE EMBODIMENTS



[0019] Reducing the overall amount of overhead and time that is needed to store or otherwise rewrite updated pages may enable an overall system, e.g., an overall host system with an embedded non-volatile memory chip, to operate more efficiently. Obtaining a new physical block each time a page associated with the physical block is to be updated when the page has a lower page number than other recently written pages may consume a significant amount of system resources and, as a result, may be relatively inefficient. On the other hand, enabling substantially any page to be written into the next available space in a block may result in an inordinate number of read operations to be performed in order to locate a particular page.

[0020] By grouping pages within blocks such that a group within a physical block may be obtained in lieu of new physical block in order to write a page into the physical block when there is at least one available group within the physical block, the overall efficiency associated with writing pages may generally be improved. In other words, the likelihood that a new block will have to be obtained in order to enable data to be written may be reduced because of the possibility that a group within the new block may be available. By using groups or sub-blocks, blocks in general may be allocated more efficiently.

[0021] Flash memory systems or, more generally, non-volatile memory devices which may benefit from the grouping of pages within a physical block generally include flash memory cards and chip sets. Typically, flash memory systems are used in conjunction with a host system such that the host system may write data to or read data from the flash memory systems. However, some flash memory systems include embedded flash memory and software which executes on a host to substantially act as a controller for the embedded flash memory, as will be discussed below with respect to Fig. 4c. Referring to Fig. 4a, a general host system which includes a non-volatile memory device, e.g., a CompactFlash memory card, will be described. A host or computer system 100 generally includes a system bus 104 which allows a microprocessor 108, a random access memory (RAM) 112, and input/output circuits 116 to communicate. It should be appreciated that host system 100 may generally include other components, e.g., display devices and networking device, which are not shown for purposes of illustration.

[0022] In general, host system 100 may be capable of capturing information including, but not limited to, still image information, audio information, and video image information. Such information may be captured in real-time, and may be transmitted to host system 100 in a wireless manner. While host system 100 may be substantially any system, host system 100 is typically a system such as a digital camera, a video camera, a cellular communications device, an audio player, or a video player. It should be appreciated, however, that host system 100 may generally be substantially any system which stores data or information, and retrieves data or information.

[0023] Host system 100 may also be a system which either only captures data, or only retrieves data. That is, host system 100 may be, in one embodiment, a dedicated system which stores data, or host system 100 may be a dedicated system which reads data. By way of example, host system 100 may be a memory writer which is arranged only to write or store data. Alternatively, host system 100 may be a device such as an MP3 player which is typically arranged to read or retrieve data, and not to capture data.

[0024] A non-volatile memory device 120 which, in one embodiment, is a removable non-volatile memory device, is arranged to interface with bus 104 to store information. An optional interface block 130 may allow non-volatile memory device 120 to interface indirectly with bus 104. When present, input/output circuit block 116 serves to reduce loading on bus 104, as will be understood by those skilled in the art. Non-volatile memory device 120 includes non-volatile memory 124 and an optional memory control system 128. In one embodiment, non-volatile memory device 120 may be implemented on a single chip or a die. Alternatively, non-volatile memory device 120 may be implemented on a multi-chip module, or on multiple discrete components which may form a chip set and may be used together as non-volatile memory device 120. One embodiment of non-volatile memory device 120 will be described below in more detail with respect to Fig. 4b.

[0025] Non-volatile memory 124, e.g., flash memory such as NAND flash memory, is arranged to store data such that data may be accessed and read as needed. Data stored in non-volatile memory 124 may also be erased as appropriate, although it should be understood that some data in non-volatile memory 124 may not be erasable. The processes of storing data, reading data, and erasing data are generally controlled by memory control system 128 or, when memory control system 128 is not present, by software executed by microprocessor 108. The operation of non-volatile memory 124 may be managed such that the lifetime of non-volatile memory 124 is substantially maximized by essentially causing sections of non-volatile memory 124 to be worn out substantially equally.

[0026] Non-volatile memory device 120 has generally been described as including an optional memory control system 128, i.e., a controller. Often, non-volatile memory device 120 may include separate chips for non-volatile memory 124 and memory control system 128, i.e., controller, functions. By way of example, while non-volatile memory devices including, but not limited to, PC cards, CompactFlash cards, MultiMedia cards, and Secure Digital cards include controllers which may be implemented on a separate chip, other non-volatile memory devices may not include controllers that are implemented on a separate chip. In an embodiment in which non-volatile memory device 120 does not include separate memory and controller chips, the memory and controller functions may be integrated into a single chip, as will be appreciated by those skilled in the art. Alternatively, the functionality of memory control system 128 may be provided by microprocessor 108, as for example in an embodiment in which non-volatile memory device 120 does not include memory controller 128, as discussed above.

[0027] With reference to Fig. 4b, non-volatile memory device 120 will be described in more detail in accordance with an embodiment of the present invention. As described above, non-volatile memory device 120 includes non-volatile memory 124 and may include memory control system 128. Memory 124 and control system 128, or controller, may be primary components of non-volatile memory device 120, although when memory 124 is an embedded NAND device, for example, non-volatile memory device 120 may not include control system 128. Memory 124 may be an array of memory cells formed on a semiconductor substrate, wherein one or more bits of data are stored in the individual memory cells by storing one of two or more levels of charge on individual storage elements of the memory cells. A non-volatile flash electrically erasable programmable read only memory (EEPROM) is an example of a common type of memory for such systems.

[0028] When present, control system 128 communicates over a bus 15 to a host computer or other system that is using the memory system to store data. Bus 15 is generally a part of bus 104 of Fig. 4a. Control system 128 also controls operation of memory 124, which may include a memory cell array 11, to write data provided by the host, read data requested by the host and perform various housekeeping functions in operating memory 124. Control system 128 generally includes a general purpose microprocessor which has associated non-volatile software memory, various logic circuits, and the like. One or more state machines are often also included for controlling the performance of specific routines.

[0029] Memory cell array 11 is typically addressed by control system 128 or microprocessor 108 through address decoders 17. Decoders 17 may apply the correct voltages to gate and bit lines of array 11 in order to program data to, read data from, or erase a group of memory cells being addressed by the control system 128. Additional circuits 19 may include programming drivers that control voltages applied to elements of the array that depend upon the data being programmed into an addressed group of cells. Circuits 19 may also include sense amplifiers and other circuits necessary to read data from an addressed group of memory cells. Data to be programmed into array 11, or data recently read from array 11, may be stored in a buffer memory 21 which, as shown, may be located within control system 128. Control system 128 also may contain various registers for temporarily storing command and status data, and the like.

[0030] Array 11 may be divided into a large number of BLOCKS 0 - N memory cells. As is common for flash EEPROM systems, a block may be the smallest unit of erase. That is, each block may contain the minimum number of memory cells that are erased together. Each block is typically divided into a number of pages. As will be appreciated by those skilled in the art, a page may be the smallest unit of programming. That is, a basic programming operation writes data into or reads data from a minimum of one page of memory cells. One or more sectors of data are typically stored within each page. As shown in Fig. 4b, one sector includes user data and overhead data. Overhead data typically includes an error correction code (ECC) that has been calculated from the user data of the sector. A portion 23 of the control system 128 may calculate the ECC when data is being programmed into array 11, and may also check the ECC when data is being read from array 11. Alternatively, the ECCs are stored in different pages, or different blocks, than the user data to which they pertain.

[0031] A sector of user data is typically 512 bytes, corresponding to the size of a sector in magnetic disk drives. Overhead data, in the described embodiment, is typically an additional 16 bytes. One sector of data is most commonly included in each page but two or more sectors may instead form a page. Any number of pages may generally form a block. By way of example, a block may be formed from eight pages up to 512, 1024 or more pages. The number of blocks is chosen to provide a desired data storage capacity for the memory system. Array 11 is typically divided into a few sub-arrays (not shown), each of which contains a proportion of the blocks, which operate somewhat independently of each other in order to increase the degree of parallelism in the execution of various memory operations. An example of the use of multiple sub-arrays is described in U.S. Patent No. 5,890,192.

[0032] In one embodiment, non-volatile memory is embedded into a system, e.g., a host system. Fig. 4c is a diagrammatic representation of a host system which includes an embedded non-volatile memory. A host or computer system 150 generally includes a system bus 154 which allows a microprocessor 158, a RAM 162, and input/output circuits 166, among other components (not shown) of host system 150, to communicate.
A non-volatile memory 174, e.g., a flash memory, allows information to be stored within host system 150. An interface 180 may be provided between non-volatile memory 174 and bus 154 to enable information to be read from and written to non-volatile memory 174.

[0033] Non-volatile memory 174 may be managed by microprocessor 158 which effectively executes either or both software and firmware which is arranged to control non-volatile memory 174. That is, microprocessor 158 may run code devices (not shown), i.e., software code devices or firmware code devices, which allow non-volatile memory 174 to be controlled. Such code devices, which may be a flash memory packaged with CPU inside microprocessor 158, a separate flash ROM, or inside non-volatile memory 174, which will be described below, may enable physical blocks in non-volatile memory 174 to be addressed, and may enable information to be stored into, read from, and erased from the physical blocks.

[0034] In general, when a user writes data, the user effectively writes data using a file system. The file system associates the data with a logical block or, more specifically, pages of a logical block, that is mapped to a physical block associated with a storage media. Fig. 5 is a diagrammatic representation of a file system with logical blocks and a media with physical blocks in accordance with an embodiment of the present invention. Logical blocks 510 of a file system 514 include any number of pages 518. The number of pages 518 included in a logical block, e.g., logical block 510a, depends on the size of an erase unit. For example, approximately thirty-two pages may be included in logical block 510a, as shown, when the smallest erase unit contains approximately thirty-two pages.

[0035] Physical blocks 530 of a media 534, e.g., an embedded flash memory, include a number of pages 538. As will be appreciated by those skilled in the art, the number of pages 538 included in physical blocks 530 is typically the same as the number of pages 518 included in logical blocks 530.

[0036] Using a data manager or data management software, as will be discussed below with respect to Fig. 5, logical blocks 510 may be mapped to physical blocks 530. Mappings between logical blocks 510 and physical blocks 530 may each be a one-to-one mapping, i.e., a mapping in which a first logical block 510 at a logical block address is mapped to a first physical block 530a at a physical block address which is substantially the same as the logical block address. However, mappings between logical blocks 510 and physical blocks 530 are often such that first logical block 510a at a logical block address is not mapped to first physical block 530a and is, instead, mapped to a different physical block, e.g., physical block 530c, which has a physical block address that is not the same as the logical block address.

[0037] Although a set of page addresses associated with a logical block may effectively all correspond to a set of page addresses associated physical block, e.g., a logical block may be associated with logical pages 0-31 while a physical block may be associated with physical pages 0-31, the contents associated with a given logical page address may not necessarily be located at a corresponding physical page address. In order to facilitate access to desired contents contained associated with a logical block and physically contained in a physical block, pages associated with a block may be grouped into a plurality of groups or sub-blocks. Either or both pages associated with a logical block and pages associated with a physical block may be grouped into groups or units. Fig. 6 is a diagrammatic representation of a block with pages which are arranged into groups in accordance with an embodiment of the present invention. A block 600 includes pages 604. It should be appreciated that block 600 may generally include any number of pages 604.

[0038] Pages 604 are arranged into groups 608. In general, both the number of groups 608 within block 600, and the number of pages 604 within a group 608, may vary widely. Typically, the number of pages 604 within a group 608 is chosen such that substantially any cluster size may be accommodated. For example, in some systems, a cluster size may be approximately four pages 604. When a cluster size is approximately four pages 604, the number of pages 604 within a group 608 may be a multiple of four, e.g., there may be four or eight pages 604 per group 608.

[0039] As will be appreciated by those skilled in the art, a cluster is generally considered to be the minimum size for a file. Hence, even when a file system attempts to write substantially only one byte for a file, the entire cluster is effectively assigned to that file. A subsequent file will typically be written into other clusters.

[0040] Returning to Fig. 6, in the embodiment as shown, each group 608 includes four pages 604. The order of pages 604 within each group 608 is fixed relative to the beginning of each group 608. By way of example, the order of pages 604a-d within group 608a is fixed, as is the order of pages 604e-h of group 608b. While the order of pages 604 within each group 608 is fixed relative to the beginning of each group 608, the order of groups 608 within block 600 is generally not fixed.

[0041] Each logical page and physical page generally includes a redundant area and a data area. In the described embodiment, a redundant area or overhead area may include up to approximately sixteen bytes of information which includes, but is not limited to, a group identifier which may be a byte which identifies the group which a particular page belongs to. Fig. 7 is a diagrammatic representation of a page in accordance with an embodiment of the present invention. A page 700 includes a redundant area 706 and a data area 710. In general page 700 includes approximately 528 bytes, of which approximately sixteen bytes may be included in redundant area 706 and approximately 512 bytes may be included in data area 710. Although redundant area 706 may be associated with the last bytes in page 700, in the described embodiment, redundant area 706 is associated with the first bytes of page 700.

[0042] Data area 710 is arranged to store data contents 712, e.g., user data contents. Redundant area 706 is arranged to include information which may enable page 700 to be tracked. For example, redundant area 706 may include a group identifier 714 which identifies which group, i.e., group within a block, page 700 is a part of. In order to facilitate the identification of a group, group identifier 714 may be the first byte associated with redundant area 706 and, hence, page 700.

[0043] Typically, when page 700 is not to be read unless group identifier 714 matches an identifier of a particular group, group identifier 714 is read and a check is made, e.g., using software or firmware within an overall system, to determine whether group identifier 714 matches the identifier associated with a particular group. Since group identifier 714 is the first byte associated with page 700, reading group identifier 714 requires relatively little overhead, and is relatively efficient. If group identifier 714 matches the identifier for a particular group, then data 712 may be read, in addition to the remainder of redundant area 706. On the other hand, if group identifier 714 does not match the identifier for the particular group, then data 712 and the remainder of redundant area 706 are not read.

[0044] By substantially only reading all bits associated with page 700 when group identifier 714 matches the identifier of a particular group, when group identifier 714 does not match the identifier of a particular group, the transfer time which would typically be associated with the transfer of approximately 527 bytes of data may be substantially eliminated. In other words, when group identifier 714 does not match the identifier of a particular group, then the time associated with reading or transferring the remaining bytes in page 700 may effectively be eliminated.
Alternatively, if the remaining bytes are to be read, e.g., because group identifier 714 matches the identifier of a particular group, then one access time, or the time which elapses between issuing a command to read data and the data being ready for transfer from the flash memory, is effectively saved. Redundant area 706 may also be placed after data area 710. Although a read operation may not be as optimized when redundant area 706 is placed after data area 710, the performance of a write operation is still generally enhanced due to a lower number of block erase operations, as described earlier.

[0045] Fig. 8a is a diagrammatic representation of a logical block and a physical block which are each divided into groups in accordance with an embodiment of the present invention. Logical block 802 and physical block 806 are evenly divided into groups 810 and groups 814, respectively. In one embodiment, the number of groups 810, 814 is typically a power of two. For example, there may be one, two, four, or eight groups 810 in logical block 802.

[0046] Logical block 802 and physical block 806 each include thirty-two pages, although it should be understood that the number of pages in logical block 802 and physical block 806 may vary widely. Each group 810 in logical block 802 includes eight pages, e.g., group 810c includes eight pages 826, and each group 814 in physical block 806 includes eight pages, e.g., group 814a includes eight pages 834. Typically, the page order in each group 810, 814 is substantially fixed. By way of example, within group 810c, a first page 826a is substantially fixed as "page 0" and an eight page 826h is substantially fixed as "page 7." As such, writing overall "page 30" of physical block 806 may be substantially equivalent to writing "page 6" 846g of group 814d, when groups 814 are ordered such that group 814a is "group 0" and group 814d is "group 3."

[0047] In general, the order of groups 814 within physical block 806 may be random. The actual group number of a group 814 may be stored in the redundant area of a page, as discussed above with respect to Fig. 7. Group numbers stored in pages 834 of group 814a are arranged to identify pages 834 as being included in group 814a. Similarly, group numbers stored in pages 838 identify pages 838 as being included in group 814b, group numbers stored in pages 842 identify pages 842 as being included in group 814c, and group numbers stored in pages 846 identify pages 846 as being included in group 814d.

[0048] As shown, the data area of pages 834 in physical group 814a, i.e., physical "group 0", are arranged to store data associated with logical group 810d. Pages 842 of physical group 814c are also arranged to store data associated with logical group 810d. The data associated with logical group 810c is arranged to be stored in the data area of pages 838 of physical group 814b. Specifically, in the described embodiment, data associated with logical group 810d is written into physical group 814a, then data associated with logical group 810c is written into physical group 814b. After physical group 814b is written to, then data associated with logical group 810d is written again, this time into physical group 814c.

[0049] With respect to Fig. 8b, one example of how physical groups 814 of physical block 806 may be written to such that group associations are as shown in Fig. 8a will be described in accordance with an embodiment of the present invention. A user may elect to write data associated with logical "page 1" 830b of logical group 810d into physical group 814a or, more specifically, physical "page 1" 834b of physical group 814a, as indicated at 860. Once physical "page 1" 834b is written into, the user may write data associated with logical "page 3" 830d into physical "page 3" 834d of physical group 814a, as indicated at 862.

[0050] If the next user command is to write data associated with logical group 810c, then even when additional pages 834 are available within physical group 814a, the data associated with logical group 810c is written into a substantially empty physical group 814, or a physical group 814 which is in a substantially erased state. For example, data associated with logical "page 2" 826c of group 810c may be written into physical "page 2" 838c of group 814b, as indicated at 864. Other pages associated with logical group 810c may be written into group 814b, as indicated at 866 and 868.

[0051] Once data associated with logical block 810c is written, the user may decide to write data associated with logical block 810d once again. As such, data associated with logical "pages 5-7" 830f-h of group 810d may be written into appropriate physical pages 842 of group 814c, as indicated at 870, 872, 874. Typically, data associated with logical "pages 5-7" 830f-h may not be written into physical group 814a because of an out-of-sequence write violation which may occur due to the fact that physical group 814b has already be written to and writing to group 814a after group 814b would generally not be allowed. As a result, data associated with logical group 810d may be located in physical group 814a and physical group 814c.

[0052] Data associated with higher overall logical block addresses associated with logical block 802 may be stored in physical pages of physical block 806 with lower overall physical block addresses than data associated with lower overall logical block addresses. By way of example, the overall logical block address associated with logical page 830b may higher than the overall logical block address associated with logical page 826c, although contents associated with logical page 830b may be written into physical page 834b which may have a lower overall physical block address than physical page 838c into which the contents associated with logical page 826c may be written.

[0053] In the event that a previously written page is overwritten, e.g., data associated with logical "page 1" 830b is written into physical group 814d (not shown) at some time after logical "page 1" 830b has been written into "page 1" 834 and physical group 814b has been written into, then the contents stored in the data area of "page 1" 834b may be considered to be invalid. As such, a command to retrieve the contents associated with logical "page 1" 830b may cause physical block 806 to effectively be searched backwards until a corresponding physical page is found. In other words, physical group 814d may be searched first, followed by physical group 814c, physical group 814b, and physical group 814a. The backwards searching will generally allow the most current data to be accessed.

[0054] Physical group 814d remains substantially unassigned, as shown, after physical groups 814a-c have been written into. In the event that logical group 810b is to be written, data associated with logical group 810b may be written into physical group 814d. That is, physical group 814d may effectively be assigned to logical group 810b, as shown in Fig. 8c. Data associated with logical group 810b is written into physical group 814d, while data associated with logical group 810c is written into physical group 814b. Logical group 810d is associated with both physical group 814a and physical group 814c.

[0055] When physical group 814d is assigned to logical group 810b, and data associated with logical group 810b is written into data areas of physical group 814d, then if data associated with logical group 810d is to be written again, there are no available physical groups 814 into which data may be written. As such, a new, e.g., spare, physical block is effectively obtained to be the physical block which corresponds to logical block 802. In other words, a new physical block is allocated such that substantially all current data associated with physical block 806, e.g., the current physical block, and new user data may be merged into the new physical block.

[0056] Fig. 8d is a diagrammatic representation of the associations between a logical block, i.e., logical block 802 of Figs. 8a-c, and a new physical block after a merge operation in accordance with an embodiment of the present invention. When the contents of physical block 806 of Fig. 8c and new user data associated with logical block 802 are merged into a new physical block 880, it should be understood that empty and invalid pages, as for example pages which have effectively been overwritten, are not written into new physical block 880. In addition, new data such as data which prompted new physical block 880 to be allocated because of a lack of available space in physical block 806 of Fig. 8c effectively overrides existing data which corresponds to the new data, i.e., new data overrides existing data which is associated with the same logical page as the new data.

[0057] Once a merge has occurred and new data has been written into new physical block 880, the association between logical block 802 and new physical block 880 may be such that data associated with group 810b of logical block 802 is stored in a physical group 884a, data associated with group 810c of logical block 802 is stored in a physical group 884b, and data associated with group 810d of logical block 802 is stored in a physical group 884c. A physical group 884d remains in a substantially erased state until physical group 884d is needed. As will be appreciated by those skilled in the art, once data is merged into or otherwise written into new physical block 880, original physical block 806 of Fig. 8a may be erased and made available for use as a spare physical block.

[0058] In order to enable read operations which access group identifiers associated with pages to occur more efficiently, a current working block information table may be built, e.g., in RAM 162 of Fig. 4c, when a page in a current block is read or written. Fig. 9 is a diagrammatic representation of a current working block information table in accordance with an embodiment of the present invention. An information table 950 is arranged to store group numbers 954. When a page is read, its group number 954 is stored into information table 950. Storing group number 954 typically serves to reduce the number of reads for the next read request of the same block, i.e., the block which includes the page whose group number 954 is stored in information table 950. Typically, substantially information pertaining to a current working block is maintained in information table 950. In order to substantially minimize the RAM requirements within an overall system, information table 950 may be built only for the block currently being accessed. When a different block is accessed, the information in an existing information table is effectively replaced by the information of the new block.

[0059] With reference to Figs. 10a and 10b, the steps associated with one method of writing contents associated with a logical page which is part of a logical group into a physical page which is part of a physical group will be described in accordance with an embodiment of the present invention. A process 1000 of writing contents into a physical page begins at step 1004 in which contents which are to be written into a physical page are identified. That is, new data which is to be written and is associated with a logical page, e.g., logical page "A," is obtained or otherwise identified.

[0060] Once contents that are to be written are identified, a determination is made in step 1008 regarding whether there is a physical group which corresponds to logical page "A." In other words, it is determined if logical page "A" or, more generally, the logical group which includes logical page "A" is the same as the last assigned or associated physical group. It if is determined that it is the last assigned physical group of logical page "A,' then a determination is made in step 1012 as to whether there is a suitable page available in the assigned physical group into which the contents associated with logical page "A" may be written. A determination of whether there is a suitable page may include a determination of whether there would be an out-of sequence write violation if the contents associated with logical page "A" were to be written into the assigned physical group. The determination may also include determining whether there is an unwritten page in the assigned physical group which is available for the contents associated with logical page "A."

[0061] If it is determined that there is a suitable page available in the assigned physical group in step 1012, then the data or contents associated with logical page "A" are written into the suitable page in step 1016. It should be appreciated that substantially any suitable method may be used to write data into a physical page. After the data is written, the process of writing data into a physical page of a physical group is completed

[0062] Alternatively, if it is determined in step 1012 that there is no suitable page available in the assigned physical group, then the indication may be that pages in the assigned physical are full, or that there are not suitable pages available that would not cause an out-of-sequence write violation. As such, process flow moves from step 1012 to step 1020 in which it is determined if there is an available physical group in the physical block. In other words, it is determined in step 1020 whether there is a group within the physical block which is not yet associated with a particular logical group.

[0063] When it is determined that there is an available physical group, then the available physical group is effectively assigned to the logical group which includes logical page "A" in step 1024. Once the available physical group is assigned to the logical group, the data or contents associated with logical page "A" is written into a suitable page in the newly assigned physical group in step 1028. Once the data is written, the process of writing data into a physical page of a physical group is completed.

[0064] Returning to step 1020 and the determination of whether there is an available physical group in the physical block assigned to logical page "A," if it is determined that there is no available physical group, then the implication is that the data or contents associated with logical page "A" may not be written into the physical block, i.e., the current physical block. Accordingly, process flow proceeds to step 1032 in which it is determined if there is a spare physical block available. If it is determined that there is no spare physical block available, then an exception is thrown in step 1036 which indicates that the contents associated with logical page "A" may not be written. Once the exception is thrown, the process of writing data into a physical page of a physical group is terminated, and a failing indication is typically reported. The design of the controller software and firmware will normally prevent this situation from substantially ever happening unless the flash memory is utilized to its end of life evenly.

[0065] On the other hand, if it is determined in step 1032 that there is an available spare physical block, the spare physical block is allocated for use with the logical block which includes logical page "A" in step 1040. Then, in step 1044, valid data from the "old" physical block that was previously assigned to the logical block which includes logical page "A" is merged into groups within the "new" or spare physical block. In addition, the new data, or the contents associated with logical page "A" which were obtained in step 1004, is written into a page of a group associated with the spare physical block. After the valid data from the old physical block is merged into the spare physical block, the old physical block is erased in step 1048. Erasing the old physical block may enable the old physical block to be subsequently used as a spare physical block, e.g., if the old physical block has not substantially reached the end of its life. Once the old physical block is erased, the process of writing data into a physical page of a physical group is terminated.

[0066] In general, the functionality associated with grouping pages within blocks is provided in software, e.g., as program code devices, or as firmware to a host system. One embodiment of a suitable system architecture associated with the software or firmware provided to a host system to enable wear leveling to occur is shown in Fig. 11. A system architecture 900 generally includes a variety of modules which may include, but are not limited to, an application interface module 904, a system manager module 908, a data manager module 912, a data integrity manager 916, and a device manager and interface module 920. In general, system architecture 900 may be implemented using software code devices or firmware which may be accessed by a processor, e.g., processor 108 of Fig. 4a.

[0067] In general, application interface module 904 may be arranged to communicate with the host, operating system or the user directly. Application interface module 904 is also in communication with system manager module 908 and data manager module 912. When the user wants to read, write or format a flash memory, the user sends requests to the operating system, the requests are passed to the application interface module 904. Application interface module 904 directs the requests to system manager module 908 or data manager module 912 depending on the requests.

[0068] System manager module 908 includes a system initialization submodule 924, an erase count block management submodule 926, and a power management block submodule 930. System initialization submodule 924 is generally arranged to enable an initialization request to be processed, and typically communicates with erase count block management submodule 926. Erase count block management submodule 926 includes functionality to cause erase counts of blocks to be stored, and functionality to cause an average erase count to be calculated, as well as updated, using individual erase counts. The use of erase counts is described in co-pending U.S. Patent Application No. 10/281,739, filed October 28, 2002 (Atty. Docket No. SANDP023).

[0069] In addition to being in communication with application interface module 904, system manager module 908 is also in communication with data manager module 812, as well as device manager and interface module 920. Data manager module 912, which communicates with both system manager module 908 and application interface module 904, may include functionality to provide sector mapping which effectively translates logical sectors into physical sectors. That is, data manager module 912 is arranged to map logical blocks into physical blocks. Data manager module 912 may also include functionality associated with operating system and file system interface layers, and enables groups within blocks to be managed.

[0070] Device manager and interface module 920, which is in communication with system manager module 908, data manager 912, and data integrity manager 916, typically provides a flash memory interface, and includes functionality associated with hardware abstractions, e.g., an I/O interface. Data integrity manager module 916 provides ECC handling, among other functions. Although only a few embodiments of the present invention have been described, it should be understood that the present invention may be embodied in many other specific forms without departing from the scope of the present invention as defined in the claims. By way example, while logical groups within logical blocks and corresponding physical groups within physical blocks have been described as being substantially the same size, logical groups and physical groups may generally be of different sizes. In other words, the number of pages included in a logical group and the number of pages included in a physical group that corresponds to the logical group may not necessarily be the same.

[0071] Therefore, the present examples are to be considered as illustrative and not restrictive, and the invention is not to be limited to the details given herein, but may be modified within the scope of the appended claims.


Claims

1. A method for writing data into a first block associated with a non-volatile memory, the first block including a plurality of pages, the plurality of physical pages being grouped into a plurality of physical groups, wherein each physical group of the plurality of physical groups includes two or more physical pages, the method comprising:

identifying a first set of data associated with a first logical page, the first logical page being one of a plurality of logical pages included in a first logical group of a logical block;

determining whether a first physical group of the plurality of physical groups is associated with the first logical group;

writing the first set of data into the first physical page responsive to determining that the first physical page is available to accommodate the first set of data;

determining whether a second physical group in the first physical block is available responsive to determining that the first physical group is not associated with the first logical group;

associating the second physical group with the first logical group responsive to determining that the second physical group is available; and

then writing the first set of data into a second physical page associated with the second physical group.


 
2. A method according to claim 1 further including:

identifying a second set of data associated with the first logical group; and writing the second set of data into a third physical page associated with the first physical group;

determining when the third physical page is available to accommodate the second set of data, wherein the second set of data is written into the third physical page when it is determined that the third physical page is available to accommodate the second set of data.


 
3. A method according to claim 1 further including:

determining whether a second physical group included in the plurality of physical groups in the first physical block is available responsive to determining that the first physical page is not available to accommodate the first set of data; and

writing the first set of data into a second physical page associated with the second physical group; and

associating the second physical group with the first logical group when it is determined that the second physical group is available.


 
4. A method according to claim 3 further including:

determining whether a second physical block associated with the non-volatile memory is available to accommodate the first set of data responsive to determining that the second physical group is not available;

copying some of the contents of the plurality of physical pages included in the first physical block into physical pages included in the second physical block; and

writing the first set of data into a third physical page associated with the second physical block.
 
5. A method according to claim 4 further including:

erasing the first physical block.


 
6. A method according to claim 4 wherein the third physical page associated with the second physical block is a third physical group included in the second physical block, the third physical group including two or more physical pages;
and further comprising;
associating the third physical group with the first logical group.
 
7. A method according to claim 1 wherein determining whether the first physical group is associated with the first logical group indicates:

obtaining a group identifier from the first physical group, wherein the group identifier is arranged to identify the first physical group; and

determining whether the group identifier is associated with the first logical group.


 
8. A system for writing data into a first physical block associated with a non-volatile memory, the first physical block including a plurality of physical pages, the plurality of physical pages being grouped into a plurality of physical groups, wherein each physical group of the plurality of physical groups includes two or more physical pages, the system comprising:

means for identifying a first set of data associated with a first logical page, the first logical page being one of a plurality of logical pages included in a first logical group of a logical block;

means for determining whether a first physical group of the plurality of physical groups is associated with the first logical group;

means for determining whether a first physical page included in the first physical group is available to accommodate the first set of data responsive to determining that the first physical group is associated with the first logical group; and

means for writing the first set of data into the first physical page responsive to determining that the first physical page is available responsive to determining that the first physical group is not associated with the first logical group;

means for associating the second physical group with the first logical group responsive to determining that the second physical group is available; and

means for then writing the first set of data into a second physical page associated with the second physical group.


 
9. A system according to claim 8, wherein the identifying means, determining means, writing means, and associating means, comprise:

a memory for storing code devices for performing the identifying, determining, writing and associating functions; and

a processor for processing the code devices.


 
10. A system according to claim 8, wherein the identifying means, determining means, writing means, and associating means, are arranged in:

a data manager module in a system architecture.


 
11. A system according to claim 8 further including:

means for identifying a second set of data associated with the first logical group;
means for writing the second set of data into a third physical page associated with the first physical group;
means for determining when the third physical page is available to accommodate the second set of data, wherein the second set of data is written into the third physical page when it is determined that the third physical page is available to accommodate the second set of data.


 
12. A system according to claim 8 further including:

means for determining whether a second physical group included in the plurality of physical groups in the first physical block is available responsive to determining that the first physical page is not available to accommodate the first set of data; and

means for writing the first set of data into a second physical page associated with the second physical group.


 
13. A system according to claim 12 further including:

means for associating the second physical group with the first logical group when it is determined that the second physical group is available.


 
14. A system according to claim 12 further including:

means for determining whether a second physical block associated with the non-volatile memory is available to accommodate the first set of data responsive to determining that the second physical group is not available.


 
15. A system according to claim 14 further including:

means for copying some of the contents of the plurality of physical pages included in the first physical block into physical pages included in the second physical block when it is determined that the second physical block is available; and

means for writing the first set data into a third physical page associated with the second physical block responsive to determining that the second physical block is available.


 
16. A system according to claim 15 wherein the third physical page associated with the second physical block is a part of a third physical group included in the second physical block, the third physical group including two or more physical pages.
 
17. A system according to claim 16 further including:

means for associating the third physical group with the first logical group.


 
18. A system according to claim 8 wherein the means for determining when the first physical group is associated with the first logical group include means for obtaining a group identifier from the first physical group, wherein the group identifier is arranged to identify the first physical group, and means for determining whether the group identifier is associated with the first logical group.
 


Ansprüche

1. Verfahren zum Einschreiben von Daten in einen mit einem nichtflüchtigen Speicher assoziierten ersten Block, wobei der erste Block eine Vielzahl von Seiten enthält, wobei die Vielzahl von physikalischen Seiten in eine Vielzahl von physikalischen Gruppen gruppiert ist, worin jede physikalische Gruppe der Vielzahl von physikalischen Gruppen zwei oder mehr physikalische Seiten enthält, wobei das Verfahren umfasst:

Identifizieren einer mit einer ersten logischen Seite assoziierten ersten Datenmenge, wobei die erste logische Seite eine einer Vielzahl von logischen Seiten ist, die in einer ersten logischen Gruppe eines logischen Blocks enthalten sind;

Ermitteln, ob eine erste physikalische Gruppe der Vielzahl von physikalischen Gruppen mit der ersten logischen Gruppe assoziiert ist;

Einschreiben der ersten Datenmenge in die erste physikalische Seite als Reaktion auf die Ermittlung, dass die erste physikalische Seite verfügbar ist, die erste Datenmenge aufzunehmen;

Ermitteln, ob eine zweite physikalische Gruppe im ersten physikalischen Block verfügbar ist, als Reaktion auf die Ermittlung, dass die erste physikalische Gruppe nicht mit der ersten logischen Gruppe assoziiert ist;

Assoziieren der zweiten physikalischen Gruppe mit der ersten logischen Gruppe als Reaktion auf die Ermittlung, dass die zweite physikalische Gruppe verfügbar ist; und

dann das Einschreiben der ersten Datenmenge in eine mit der zweiten physikalischen Gruppe assoziierte zweite physikalische Seite.


 
2. Verfahren nach Anspruch 1, außerdem umfassend:

Identifizieren einer mit der ersten logischen Gruppe assoziierten zweiten Datenmenge; und Einschreiben der zweiten Datenmenge in eine mit der ersten physikalischen Gruppe assoziierte dritte physikalische Seite;

Ermitteln, wann die dritte physikalische Seite verfügbar ist, die zweite Datenmenge aufzunehmen, worin die zweite Datenmenge in die dritte physikalische Seite eingeschrieben wird, wenn ermittelt wird, dass die dritte physikalische Seite verfügbar ist, die zweite Datenmenge aufzunehmen.


 
3. Verfahren nach Anspruch 1, außerdem umfassend:

Ermitteln, ob eine in der Vielzahl von physikalischen Gruppen im ersten physikalischen Block enthalte zweite physikalische Gruppe verfügbar ist, als Reaktion auf die Ermittlung, dass die erste physikalische Seite nicht verfügbar ist, die erste Datenmenge aufzunehmen; und

Einschreiben der ersten Datenmenge in eine mit der zweiten physikalischen Gruppe assoziierte zweite physikalische Seite; und

Assoziieren der zweiten physikalischen Gruppe mit der ersten logischen Gruppe, wenn ermittelt wird, dass die zweite physikalische Gruppe verfügbar ist.


 
4. Verfahren nach Anspruch 3, außerdem umfassend:

Ermitteln, ob ein mit dem nichtflüchtigen Speicher assoziierte zweiter physikalischer Block verfügbar ist, die erste Datenmenge aufzunehmen, als Reaktion auf die Ermittlung, dass die zweite physikalische Gruppe nicht verfügbar ist;

Kopieren eines Teils des Inhalts der im ersten physikalischen Block enthaltenen Vielzahl von physikalischen Seiten in physikalische Seiten, die im zweiten physikalischen Block enthalten sind; und

Einschreiben der ersten Datenmenge in eine mit dem zweiten physikalischen Block assoziierte dritte physikalische Seite.


 
5. Verfahren nach Anspruch 4, außerdem umfassend:

Löschen des ersten physikalischen Blocks.


 
6. Verfahren nach Anspruch 4, worin die mit dem zweiten physikalischen Block assoziierte dritte physikalische Seite eine dritte physikalische Gruppe ist, die im zweiten physikalischen Block enthalten ist, wobei die dritte physikalische Gruppe zwei oder mehr physikalische Seiten enthält;
und außerdem umfassend:

Assoziieren der dritten physikalischen Gruppe mit der ersten logischen Gruppe.


 
7. Verfahren nach Anspruch 1, worin die Ermittlung, ob die erste physikalische Gruppe mit der ersten logischen Gruppe assoziiert ist, Folgendes anzeigt:

Erhalten eines Gruppenidentifikators von der ersten physikalischen Gruppe, worin der Gruppenidentifikator dazu angeordnet ist, die erste physikalische Gruppe zu identifizieren; und

Ermitteln, ob der Gruppenidentifikator mit der ersten logischen Gruppe assoziiert ist.


 
8. System zum Einschreiben von Daten in einen mit einem nichtflüchtigen Speicher assoziierten ersten physikalischen Block, wobei der erste physikalische Block eine Vielzahl von physikalischen Seiten enthält, wobei die Vielzahl von physikalischen Seiten in eine Vielzahl von physikalischen Gruppen gruppiert ist, worin jede physikalische Gruppe der Vielzahl von physikalischen Gruppen zwei oder mehr physikalische Seiten enthält, wobei das System umfasst:

Mittel zum Identifizieren einer mit einer ersten logischen Seite assoziierten ersten Datenmenge, wobei die erste logische Seite eine einer Vielzahl von logischen Seiten ist, die in einer ersten logischen Gruppe eines logischen Blocks enthalten ist;

Mittel zum Ermitteln, ob eine erste physikalische Gruppe der Vielzahl von physikalischen Gruppen mit der ersten logischen Gruppe assoziiert ist;

Mittel zum Ermitteln, ob eine in der ersten physikalischen Gruppe enthaltene erste physikalische Seite zum Aufnehmen der ersten Datenmenge verfügbar ist, als Reaktion auf die Ermittlung, dass die erste physikalische Gruppe mit der ersten logischen Gruppe assoziiert ist; und

Mittel zum Einschreiben der ersten Datenmenge in die erste physikalische Seite als Reaktion auf die Ermittlung, dass die erste physikalische Seite verfügbar ist, als Reaktion auf die Ermittlung, dass die erste physikalische Gruppe nicht mit der ersten logischen Gruppe assoziiert ist;

Mittel zum Assoziieren der zweiten physikalischen Gruppe mit der ersten logischen Gruppe als Reaktion auf die Vermittlung, dass die zweite physikalische Gruppe verfügbar ist; und

Mittel, um dann die erste Datenmenge in eine mit der zweiten physikalischen Gruppe assoziierte zweite physikalische Seite einzuschreiben.


 
9. System nach Anspruch 8, worin die Identifikationsmittel, Ermittlungsmittel, Schreibmittel und Assoziationsmittel umfassen:

einen Speicher zum Speichern von Codiereinrichtungen, um die Identifikations-, Ermittlungs-, Schreib- und Assoziationsfunktionen auszuführen; und

einen Prozessor zum Verarbeiten der Codiereinrichtungen.


 
10. System nach Anspruch 8, worin die Identifikationsmittel, Ermittlungsmittel, Schreibmittel und Assoziationsmittel angeordnet sind in:

einem Datenverwaltungsmodul in einer Systemarchitektur.


 
11. System nach Anspruch 8, außerdem umfassend:

Mittel zum Identifizieren einer zweiten Datenmenge, die mit der ersten logischen Gruppe assoziiert ist;

Mittel zum Einschreiben der zweiten Datenmenge in eine mit der ersten physikalischen Gruppe assoziierte dritte physikalische Seite;

Mittel zum Ermitteln, wann die dritte physikalische Seite verfügbar ist, die zweite Datenmenge aufzunehmen, worin die zweite Datenmenge in die dritte physikalische Seite eingeschrieben wird, wenn ermittelt wird, dass die dritte physikalische Seite verfügbar ist, die zweite Datenmenge aufzunehmen.


 
12. System nach Anspruch 8, außerdem umfassend:

Mittel zum Ermitteln, ob eine in der Vielzahl von physikalischen Gruppen im ersten physikalischen Block enthaltene zweite physikalische Gruppe verfügbar ist, als Reaktion auf die Ermittlung, dass die erste physikalische Seite nicht verfügbar ist, die erste Datenmenge aufzunehmen; und

Mittel zum Einschreiben der ersten Datenmenge in eine mit der zweiten physikalischen Gruppe assoziierte zweite physikalische Seite.


 
13. System nach Anspruch 12, außerdem umfassend:

Mittel zum Assoziieren der zweiten physikalischen Gruppe mit der ersten logischen Gruppe, wenn ermittelt wird, dass die zweite physikalische Gruppe verfügbar ist.


 
14. System nach Anspruch 12, außerdem umfassend:

Mittel zum Ermitteln, ob ein mit dem nichtflüchtigen Speicher assoziierte zweiter physikalischer Block verfügbar ist, die erste Datenmenge aufzunehmen, als Reaktion auf die Ermittlung, dass die zweite physikalische Gruppe nicht verfügbar ist.


 
15. System nach Anspruch 14, außerdem umfassend:

Mittel zum Kopieren eines Teils des Inhalts der im ersten physikalischen Block enthaltenen Vielzahl von physikalischen Seiten in physikalische Seiten, die im zweiten physikalischen Block enthalten sind, wenn ermittelt wird, dass der zweite physikalische Block verfügbar ist; und

Mittel zum Einschreiben der ersten Datenmenge in eine mit dem zweiten physikalischen Block assoziierte dritte physikalische Seite als Reaktion auf die Ermittlung, dass der zweite physikalische Block verfügbar ist.


 
16. System nach Anspruch 15, worin die mit dem zweiten physikalischen Block assoziierte dritte physikalische Seite ein Teil einer dritten physikalischen Gruppe ist, die im zweiten physikalischen Block enthalten ist, wobei die dritte physikalische Gruppe zwei oder mehr physikalische Seiten enthält.
 
17. System nach Anspruch 16, außerdem umfassend:

Mittel zum Assoziieren der dritten physikalischen Gruppe mit der ersten logischen Gruppe.


 
18. System nach Anspruch 8, worin die Mittel zum Ermitteln, wann die erste physikalische Gruppe mit der ersten logischen Gruppe assoziiert ist, Mittel enthalten, um von der ersten physikalischen Gruppe einen Gruppenidentifikator zu erhalten, worin der Gruppenidentifikator dazu angeordnet ist, die erste physikalische Gruppe zu identifizieren, und Mittel zum Ermitteln, ob der Gruppenidentifikator mit der ersten logischen Gruppe assoziiert ist.
 


Revendications

1. Procédé pour écrire des données dans un premier bloc associé à une mémoire non volatile, le premier bloc comprenant une pluralité de pages, la pluralité de pages physiques étant groupées dans une pluralité de groupes physiques, dans lequel chaque groupe physique de la pluralité de groupes physiques comprend deux pages physiques ou plus, le procédé consistant à :

identifier un premier ensemble de données associées à une première page logique, la première page logique étant l'une d'une pluralité de pages logiques incluses dans un premier groupe logique d'un bloc logique ;

déterminer si un premier groupe physique de la pluralité de groupes physiques est associé au premier groupe logique ;

écrire le premier ensemble de données dans la première page physique en réponse à la détermination du fait que la première page physique est disponible pour recevoir le premier ensemble de données;

déterminer si un deuxième groupe physique dans le premier bloc physique est disponible en réponse à la détermination du fait que le premier groupe physique n'est pas associé au premier groupe logique ;

associer le deuxième groupe physique au premier groupe logique en réponse à la détermination du fait que le deuxième groupe physique est disponible ; et

écrire ensuite le premier ensemble de données dans une deuxième page physique associée au deuxième groupe physique.


 
2. Procédé selon la revendication 1 consistant en outre à :

identifier un deuxième ensemble de données associées au premier groupe logique ; et écrire le deuxième ensemble de données dans une troisième page physique associée au premier groupe physique ;

déterminer quand la troisième page physique est disponible pour recevoir le deuxième ensemble de données, dans lequel le deuxième ensemble de données est écrit dans la troisième page physique lorsqu'il est déterminé que la troisième page physique est disponible pour recevoir le deuxième ensemble de données.


 
3. Procédé selon la revendication 1 consistant en outre à :

déterminer si un deuxième groupe physique inclus dans la pluralité de groupes physiques dans le premier bloc physique est disponible en réponse à la détermination du fait que la première page physique n'est pas disponible pour recevoir le premier ensemble de données ; et

écrire le premier ensemble de données dans une deuxième page physique associée au deuxième groupe physique ; et

associer le deuxième groupe physique au premier groupe logique lorsqu'il est déterminé que le deuxième groupe physique est disponible.


 
4. Procédé selon la revendication 3 consistant en outre à :

déterminer si un deuxième bloc physique associé à la mémoire non volatile est disponible pour recevoir le premier ensemble de données en réponse à la détermination du fait que le deuxième groupe physique n'est pas disponible ;

copier certains des contenus de la pluralité de pages physiques incluses dans le premier bloc physique dans des pages physiques incluses dans le deuxième bloc physique ; et

écrire le premier ensemble de données dans une troisième page physique associée au deuxième bloc physique.


 
5. Procédé selon la revendication 4 consistant en outre à :

effacer le premier bloc physique.


 
6. Procédé selon la revendication 4, dans lequel la troisième page physique associée au deuxième bloc physique est un troisième groupe physique inclus dans le deuxième bloc physique, le troisième groupe physique comprenant deux pages physiques ou plus ;
et consistant en outre à :

associer le troisième groupe physique au premier groupe logique.


 
7. Procédé selon la revendication 1, dans lequel la détermination du fait que le premier groupe physique est ou non associé au premier groupe logique indique :

l'obtention d'un identifiant de groupe du premier groupe physique, dans lequel l'identifiant de groupe est agencé pour identifier le premier groupe physique ; et

la détermination du fait que l'identifiant de groupe est ou non associé au premier groupe logique.


 
8. Système pour écrire des données dans un premier bloc physique associé à une mémoire non volatile, le premier bloc physique comprenant une pluralité de pages physiques, la pluralité de pages physiques étant groupées dans une pluralité de groupes physiques, dans lequel chaque groupe physique de la pluralité de groupes physiques comprend deux pages physiques ou plus, le système comprenant :

des moyens pour identifier un premier ensemble de données associées à une première page logique, la première page logique étant l'une d'une pluralité de pages logiques incluses dans un premier groupe logique d'un bloc logique ;

des moyens pour déterminer si un premier groupe physique de la pluralité de groupes physiques est associé au premier groupe logique ;

des moyens pour déterminer si une première page physique incluse dans le premier groupe physique est disponible pour recevoir le premier ensemble de données en réponse à la détermination du fait que le premier groupe physique est associé au premier groupe logique ; et

des moyens pour écrire le premier ensemble de données dans la première page physique en réponse à la détermination du fait que la première page physique est disponible en réponse à la détermination du fait que le premier groupe physique n'est pas associé au premier groupe logique ;

des moyens pour associer le deuxième groupe physique au premier groupe logique en réponse à la détermination du fait que le deuxième groupe physique est disponible ; et

des moyens pour écrire ensuite le premier ensemble de données dans une deuxième page physique associée au deuxième groupe physique.


 
9. Système selon la revendication 8, dans lequel les moyens d'identification, les moyens de détermination, les moyens d'écriture et les moyens d'association comprennent :

une mémoire pour mémoriser des dispositifs de code pour effectuer les fonctions d'identification, de détermination, d'écriture et d'association ; et

un processeur pour traiter les dispositifs de code.


 
10. Système selon la revendication 8, dans lequel les moyens d'identification, les moyens de détermination, les moyens d'écriture et les moyens d'association sont agencés dans :

un module de gestion de données dans une architecture de système.


 
11. Système selon la revendication 8, comprenant en outre :

des moyens pour identifier un deuxième ensemble de données associées au premier groupe logique ;

des moyens pour écrire le deuxième ensemble de données dans une troisième page physique associée au premier groupe physique ;

des moyens pour déterminer quand la troisième page physique est disponible pour recevoir le deuxième ensemble de données, dans lequel le deuxième ensemble de données est écrit dans la troisième page physique lorsqu'il est déterminé que la troisième page physique est disponible pour recevoir le deuxième ensemble de données.


 
12. Système selon la revendication 8, comprenant en outre :

des moyens pour déterminer si un deuxième groupe physique inclus dans la pluralité de groupes physiques dans le premier bloc physique est disponible en réponse à la détermination du fait que la première page physique n'est pas disponible pour recevoir le premier ensemble de données; et

des moyens pour écrire le premier ensemble de données dans une deuxième page physique associée au deuxième groupe physique.


 
13. Système selon la revendication 12, comprenant en outre :

des moyens pour associer le deuxième groupe physique au premier groupe logique lorsqu'il est déterminé que le deuxième groupe physique est disponible.


 
14. Système selon la revendication 12, comprenant en outre :

des moyens pour déterminer si un deuxième bloc physique associé à la mémoire non volatile est disponible pour recevoir le premier ensemble de données en réponse à la détermination du fait que le deuxième groupe physique n'est pas disponible.


 
15. Système selon la revendication 14, comprenant en outre :

des moyens pour copier certains des contenus de la pluralité de pages physiques incluses dans le premier bloc physique dans des pages physiques incluses dans le deuxième bloc physique lorsqu'il est déterminé que le deuxième bloc physique est disponible ; et

des moyens pour écrire le premier ensemble de données dans une troisième page physique associée au deuxième bloc physique en réponse à la détermination du fait que le deuxième bloc physique est disponible.


 
16. Système selon la revendication 15, dans lequel la troisième page physique associée au deuxième bloc physique fait partie d'un troisième groupe physique inclus dans le deuxième bloc physique, le troisième groupe physique comprenant deux pages physiques ou plus.
 
17. Système selon la revendication 16, comprenant en outre :

des moyens pour associer le troisième groupe physique au premier groupe logique.


 
18. Système selon la revendication 8, dans lequel les moyens pour déterminer quand le premier groupe physique est associé au premier groupe logique comprennent des moyens pour obtenir un identifiant de groupe du premier groupe physique, dans lequel l'identifiant de groupe est agencé pour identifier le premier groupe physique, et des moyens pour déterminer si l'identifiant de groupe est associé au premier groupe logique.
 




Drawing


















































Cited references

REFERENCES CITED IN THE DESCRIPTION



This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Patent documents cited in the description