Direct data file storage implementation techniques in flash memories

• US 7,984,233 B2
• Filed: 05/05/2010
• Issued: 07/19/2011
• Est. Priority Date: 02/16/2005
• Status: Active Grant

• Current Assignee: SanDisk Technologies LLC (Western Digital Corporation)
• Original Assignee: SanDisk Corporation (Western Digital Corporation)
• Inventors: Sinclair, Alan W.
• Primary Examiner(s): Kim; Matt
• Assistant Examiner(s): Chrzanowski; Matthew R
• Application Number:
  US12/774,109
• Publication Number:
  US 20100217926A1
• Time in Patent Office: 440 Days
• Field of Search: None
• US Class Current: 711/103
• CPC Class Codes: G06F 12/0246 in block erasable memory, e.g. flash memory
• CPC Class Codes: G06F 2212/7201 Logical to physical mapping or translation of blocks or pages
• CPC Class Codes: G06F 2212/7205 Cleaning, compaction, garbage collection, erase control
• CPC Class Codes: G06F 3/0607 by facilitating the process of upgrading existing storage systems, e.g. for improving compatibility between host and storage device
• CPC Class Codes: G06F 3/0643 Management of files
• CPC Class Codes: G06F 3/0679 Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]

Abstract

Host system data files are written directly to a large erase block flash memory system with a unique identification of each file and offsets of data within the file but without the use of any intermediate logical addresses or a virtual address space for the memory. Directory information of where the files are stored in the memory is maintained within the memory system by its controller, rather than by the host. The file based interface between the host and memory systems allows the memory system controller to utilize the data storage blocks within the memory with increased efficiency.

Claims

1. A mass storage system, comprising:

an array of non-volatile charge storage semiconductor memory cells organized into blocks that individually have a capacity of storing multiple units of received data that individually include at least 512 bytes of data, the memory cells of the individual blocks being concurrently erased prior to any new data being written therein, and

a controller connected with the memory cell array and adapted to receive data of individual files having a logical address including a unique file identification and offsets of data within an identified file to program the received data at physical addresses within the blocks of memory cells, and which further operates to maintain a plurality of records of the programmed data for the individual files that identify groups of variable amounts of data making up the identified file, data within the groups individually having both contiguous logical offset addresses and contiguous physical addresses, wherein each individual record of the plurality of records include the file identification, at least a logical offset address of data within the identified file and a corresponding physical address of the data within the group, and the controller further operates to program in a single one of the blocks of memory cells at least two of the groups of data from the identified file, wherein the at least two of the groups of data have a discontinuity of offset addresses between the at least two of the groups of data and contiguous physical addresses between the at least two of the groups of data,

the mass storage system directly translates file identifiers and offsets received from a host into physical addresses of blocks and pages of memory cells in which data of the individual files are written without the use of any intermediate logical address conversion.

2. The mass storage system of claim 1, wherein each individual record of the plurality of records additionally includes a length of the data within the group represented by each individual record.

3. The mass storage system of claim 1, wherein the memory cell array includes a number of memory cells sufficient to store at least 256 megabytes of data.

4. The mass storage system of claim 1, wherein the blocks individually contain at least 1024 bytes of data.

5. A mass storage system, comprising:

an array of non-volatile charge storage semiconductor memory cells organized into blocks that are erased prior to any new data being written therein and which individually have a capacity of storing at least 1024 bytes of received data,

a controller connected with the memory cell array and adapted to receive data from outside the mass storage system of individual files of varying amounts of data having a logical address including a unique file identification and offsets of data within an identified file to program the received data at physical addresses within the blocks of memory cells as defined groups of data individually having a size equal to or less than the capacity of the individual memory cell blocks and containing data having continuous logical offsets and continuous physical addresses,

a file map capable of storing records of individual groups of data stored in the memory cell array, individual data records for a file of data including a starting logical offset address, a corresponding starting physical address and length of the data group,

the mass storage system directly translates file identifiers and offsets received from an host into physical addresses of blocks and pages of memory cells in which data of the individual files are written without the use of any intermediate logical address conversion, and

the controller further operating to program in a single one of the blocks of memory cells at least two of the defined groups of data from the identified file, wherein the at least two of the defined groups of data have a discontinuity of offset addresses between the at least two of the groups of data and contiguous physical addresses between the at least two of the groups of data.

6. The mass storage system of claim 5, wherein the starting physical address of the individual data records includes an identification of a block of memory cells and a byte location within the identified block.

7. A computer system, comprising:

a host computing system including a processor that provides data files that are individually identified by a unique identifier and offsets of data within an identified file,

a mass data storage system including a controller and memory cells organized in separate blocks of memory cells that are concurrently erased prior to writing therein data therein and multiple pages within the individual blocks, the controller operating to write data files received from the host into the memory cells at physical addresses within the pages of the blocks as defined groups of data individually having a size equal to or less than a data storage capacity of the individual blocks and containing varying amounts of data with both continuous logical and physical addresses, the memory system having a capacity to store at least 256 megabytes of data,

a file map stored in the memory cells and managed by the controller to store records of individual groups of the data stored in the memory cells, individual records of the groups of file data includes one unique file identifier, at least one logical offset address of data within an identified file and one corresponding physical address for one group of data, without any intermediate logical address, wherein the controller uses the file map to locate physical addresses of data of a file having one unique file identifier, and

the controller further operating to program in a single one of the blocks of memory cells at least two of the groups of data from the identified file, wherein the at least two of the groups of data have a discontinuity of offset addresses between the at least two of the groups of data and contiguous physical addresses between the at least two of the groups of data.

8. The computer system of claim 7, wherein controller stores the offset and physical addresses in the records of individual groups of file data with the same resolution.

9. The computer system of claim 8, wherein the same resolution is one byte.

10. A method of operating a re-programmable non-volatile mass storage system having memory cells organized into blocks that are erased prior to any new data being written therein and which individually have a capacity of storing multiple units of data that individually include at least 512 bytes of data, comprising:

receiving data of a plurality of files of varying amounts of data that are individually identified by logical addresses including a unique file identification and offsets of data within the identified file,

mapping addresses of the received data from the received unique file identifications and offsets therein directly into physical addresses of memory cell blocks of the mass storage system and offsets therein by use of a file map, the direct mapping occurring without performing an intermediate logical address conversion,

programming the received data at the physical addresses within the blocks of memory cells to which the received data are mapped, and

maintaining a plurality of records in the file map of the programmed data for individual files that identify groups of variable amounts of data making up the file, the data of individual groups of data having both contiguous logical offset addresses and contiguous physical addresses, the plurality of records for at least one of the files including at least two records of at least two of the groups of data wherein the at least two of the groups of data have between them a discontinuity of logical offset addresses and a continuity of physical addresses, the at least two groups of data being programmed entirely within one of the blocks of memory cells.

11. The method of claim 10, wherein maintaining the plurality of records includes providing in each individual record at least a beginning logical file offset address and a corresponding beginning physical address of the data within one of the groups of data to which each individual record pertains.

12. The method of claim 11, wherein maintaining the plurality of records additionally includes providing in each individual record a length of the data within the groups to which each individual records pertain.

13. The method of claim 12, wherein maintaining the file map includes maintaining the plurality of records organized by grouping records of the individual files together.

14. The method of claim 10, wherein programming data includes programming the received data of a new file not currently stored within the mass storage system by beginning programming the data into a first physical address of an erased block and continuing through the erased block and then, if the amount of data in the new file exceeds the storage capacity of the erased block, into at least one other erased block in physical order until all the received data of the file have been programmed in one programming operation.

15. The method of claim 10, wherein programming data includes programming received additional data of an existing file previously stored in the memory system by beginning programming the data into a physical address of a block immediately after the existing file data ends in said block and continuing through an erased portion of that block and then, if the amount of additional data causes the storage capacity of that block to be exceeded, into at least one other erased block in physical order until all the additional data of the file have been programmed in one programming operation.

16. The method of claim 15, wherein programming the additional data includes programming data of an insert within the previously stored file.

17. The method of claim 16, wherein maintaining a plurality of records includes changing records of the existing data groups of the existing file and adding at least one additional record of at least one other data group.

18. The method of claim 15, wherein programming the additional data includes programming data of an update to a portion of the data of the previously stored file which results in the data portion of the previously stored file becoming obsolete.

19. The method of claim 18, wherein maintaining a plurality of records includes changing records of the existing data groups of the existing file and adding at least one additional record of at least one other data group.

20. The method of claim 19, wherein the mass storage system contains a number of memory cells sufficient to store at least 256 megabytes of data.

21. The method of claim 19, wherein the units of data individually contain at least 1024 bytes.

22. The method of claim 19, wherein at least 8 units of data are written into individual blocks of the mass storage system.

23. The method of claim 10, wherein the plurality of files received by the mass storage system include files of data generated by a host system from application programs running on the host system.

24. The method of claim 10, wherein the individual groups of data contain an amount of data that is less than a data storage capacity of the individual memory cell blocks.

25. A method of transferring data between a host system and a re-programmable non-volatile mass storage system having blocks of memory cells that are erasable together, an individual block being erased prior to new data being written therein, wherein:

the host system identifies individual files of varying amounts of data it generates by logical addresses including unique file identifiers and offsets of data within the individual files, and sends such file identifiers and offsets to the mass storage system along with data of the individual files,

the mass storage system maps the logical addresses of the received data of the individual files directly into physical addresses including an identifier of memory cell blocks of the mass storage system and offsets of data therein, and stores the data of the individual files at the physical addresses of the mass storage system, the direct mapping occurring without performing an intermediate logical address conversion, and

the mass storage system writes the data of an individual one of the individual files in an order received from the host system into at least one previously erased block of memory cells to which the file data are mapped with contiguous physical addresses independent of data offsets within the individual one of the individual files,

wherein for at least one of the individual files including at least two records of at least two groups of data wherein the at least two of the groups of data have between them a discontinuity of logical offset addresses and a continuity of physical addresses, the at least two groups of data being programmed entirely within one of the blocks of memory cells.

26. The method of claim 25 wherein the mass storage system further writes the data of an individual one of the individual files received from the host system to fill at least one memory block to which the file data are mapped, and any residual data of an amount less than enough to fill another memory block are written into a common block that also contains residual data of a different at least one of the individual files.

27. The method of claim 25, wherein the individual files of data generated by the host system are generated from application programs running on the host system.

28. The method of claim 25, wherein the mass storage system writes the data of a given file into one previously erased block of memory cells as two or more groups of data that individually have offsets that are contiguous within each group of the groups but which are non-contiguous with each other.

29. The method of claim 25, wherein the mass storage system writes the data into at least one previously erased block of memory cells in groups of variable amounts of data less than a data storage capacity of the individual memory cell blocks, the groups of data individually having continuous offsets of data within one of the individual files and continuous offsets within an addressed block of memory cells.