Partial block data programming and reading operations in a non-volatile memory

• US 6,968,421 B2
• Filed: 05/07/2004
• Issued: 11/22/2005
• Est. Priority Date: 01/19/2001
• Status: Expired due to Term

• Current Assignee: SanDisk Technologies LLC (Western Digital Corporation)
• Original Assignee: SanDisk Corporation (Western Digital Corporation)
• Inventors: Conley, Kevin M.
• Primary Examiner(s): Sparks, Donald
• Assistant Examiner(s): Dinh, Ngoc V
• Application Number:
  US10/841,388
• Publication Number:
  US 20040210708A1
• Time in Patent Office: 564 Days
• Field of Search: 711/103, 711/113, 711/115, 711/154, 365/185.11, 365/185.33, 365/218
• US Class Current: 711/103
• CPC Class Codes: G06F 12/0246 in block erasable memory, e.g. flash memory
• CPC Class Codes: G06F 2212/1016 Performance improvement
• CPC Class Codes: G06F 2212/1032 Reliability improvement, data loss prevention, degraded operation etc
• CPC Class Codes: G06F 2212/7202 Allocation control and policies
• CPC Class Codes: G06F 2212/7208 Multiple device management, e.g. distributing data over multiple flash devices
• CPC Class Codes: G06F 2212/7209 Validity control, e.g. using flags, time stamps or sequence numbers
• CPC Class Codes: G11C 16/102 External programming circuits, e.g. EPROM programmers; In-circuit programming or reprogramming; EPROM emulators
• CPC Class Codes: G11C 16/105 Circuits or methods for updating contents of nonvolatile memory, especially with 'security' features to ensure reliable replacement, i.e. preventing that old data is lost before new data is reliably written
• CPC Class Codes: G11C 2216/16 Flash programming of all the cells in an array, sector or block simultaneously

Abstract

Data in less than all of the pages of a non-volatile memory block are updated by programming the new data in unused pages of either the same or another block. In order to prevent having to copy unchanged pages of data into the new block, or to program flags into superceded pages of data, the pages of new data are identified by the same logical address as the pages of data which they superceded and a time stamp is added to note when each page was written. When reading the data, the most recent pages of data are used and the older superceded pages of data are ignored. This technique is also applied to metablocks that include one block from each of several different units of a memory array, by directing all page updates to a single unused block in one of the units.

Claims

1. A method of storing and retrieving original and replacement data in a non-volatile memory system, comprising:

identifying units of the original and the replacement data by the same logical address,

reading units of data in an inverse order from an order in which they were programmed into the memory, and

distinguishing units of replacement data from units of original data having the same logical address by the order in which they are read.

13. The method of any one of claim 1, 2, 3, or 7, additionally comprising operating storage elements of the individual memory cells with more than two storage states, thereby storing more than one bit of data in each storage element.

14. The method of claim 13, wherein the storage elements include individual floating gates.

15. The method of any one of claim 1, 2, 3, or 7, wherein the non-volatile memory system is formed within an enclosed card having an electrical connector along one edge thereof that operably connects with a host system.

2. In a non-volatile memory system having a plurality of blocks of memory storage elements that are individually organized into a plurality of pages of memory storage elements, a method of substituting new data for superceded data within at least one page of one of the plurality of blocks while data in at least another page of said one block is not replaced, comprising:

programming the new data into at least one page of said one or another of the plurality of blocks,

identifying the at least one page of superceded data and the at least one page of new data by a common logical address,

recording a relative time of programming the new and the superceded data,

wherein the relative time of programming is recorded for the individual blocks, thereby to identify an order of programming of individual blocks having data with a common logical address, and

wherein pages within the individual blocks are programmed in a designated order, whereby the new pages of data are distinguishable from superceded pages of data within an individual block by their relative positions within the block.

11. A method of reading data that has been updated according to claim 2, comprising:

reading pages of data within said one and, if new data has been programmed thereinto, another block in a reverse order from which they were programmed, and

passing over any pages of data so read which have the same logical page address as a page whose data has already been read.

3. In a non-volatile memory system having a plurality of blocks of memory storage elements that are individually organized into a plurality of pages of memory storage elements, a method of substituting new data for superceded data within at least one page of one of the plurality of blocks while data in at least another page of said one block is not being superceded, comprising:

programming the new data into at least one page of another of the plurality of blocks,

identifying the at least one page of superceded data and the at least one page of new data by a common logical address,

recording a relative time of programming the new and the superceded data, and

wherein the data in at least another page of said one block that are not superceded are not copied into said another block as part of substituting the new data for the superceded data.

4. The method of claim 3, wherein recording a relative time of programming the new and superceded data includes storing a value of a clock at each of the times that the new and superceded data are programmed.

5. The method of claim 3, wherein recording a relative time of programming the new and superceded data includes storing a different value of a sequence of numbers at each of the times that the new and superceded data are programmed.

6. The method of either of claim 4 or 5, wherein storing the value indicating a relative time of programming the new and superceded data includes storing the individual values within the same pages as the new and superceded data to which the values relate.

7. In a non-volatile memory system having a plurality of blocks of memory storage elements that are individually organized into a plurality of pages of memory storage elements, a method of substituting new data for superceded data within at least one page of one of the plurality of blocks while data in at least another page of said one block is not superceded, comprising:

programming the new data into at least one page of another of the plurality of blocks,

identifying the at least one page of superceded data and the at least one page of new data by a common logical address,

recording a relative time of programming the new and the superceded data, and

wherein programming the new data into at least one page of another of the plurality of blocks includes programming the superceded data into the first available unused pages within said another block in a predefined order.

8. The method of claim 7, wherein identifying the at least one page of superceded data and the at least one page of new data by a common logical address includes recording at least part of the common logical address in the individual pages as overhead data.

9. The method of claim 8, including building a table in volatile memory including multiple physical block addresses for the common logical address.

10. In a non-volatile memory system having a plurality of blocks of memory storage elements that are individually organized into a plurality of pages of memory storage elements, a method of substituting new data for superceded data within at least one page of one of the plurality of blocks while data in at least another page of said one block is not superceded and thereafter reading data that has been updated comprising:

programming the new data into at least one page of said one or another of the plurality of blocks,

identifying the at least one page of superceded data and the at least one page of new data by a common logical address, and

recording a relative time of programming the new and the superceded data, wherein the relative time of programming is recorded for the individual pages in which the new and superceded data are programmed, whereby the at least one page of new data is distinguishable from the at least one page of superceded data by their recorded relative times of programming,

reading pages of data from said one block and, if new data has been programmed thereinto, said another block,

identifying any multiple pages of data that have the same logical address,

utilizing the recorded relative times of programming the new and superceded data to identify the most current of any pages having the same logical address, and

assembling data in the most current of any pages having the same logical address along with pages in said at least another page of said one block that have not been superceded.

12. The method of either one of claim 10 or 11, additionally comprising operating the individual memory storage elements with more than two storage states, thereby storing more than one bit of data in each storage element, and reading pages of data includes reading the more than two storage states from the individual memory storage elements.

16. A method of operating a non-volatile memory system having an array of memory storage elements organized into at least two sub-arrays, wherein the individual sub-arrays are divided into a plurality of non-overlapping blocks of storage elements wherein a block contains the smallest group of memory storage elements that are erasable together, and the individual blocks are divided into a plurality of pages of storage elements wherein a page is the smallest group of memory storage elements that are programmable together, comprising:

linking at least one block from individual ones of said at least two sub-arrays to form a metablock,

simultaneously erasing the component blocks together as a unit, and

updating pages of original data within at least one of the component blocks of a metablock less than all the pages within the at least one block by programming replacement data into pages within another at least one block in only a designated one of the sub-arrays regardless of which sub-array the data being updated is stored.

17. The method of claim 16, wherein storing the original and replacement data includes:

identifying the original and replacement data by the same logical address to the memory system, and

distinguishing the replacement data from the original data by keeping track of the relative times that the original and replacement data have been programmed in their respective pages of the memory.

18. A non-volatile memory system, comprising:

an array of non-volatile memory storage elements organized in blocks of storage elements, wherein an individual block contains the smallest group of storage elements that is erasable,

a programming mechanism that writes into a first block an updated version of less than all of original data stored in a second block along with an indication of the later writing of the updated version,

an address mechanism that logically addresses both the original data and the updated version with the same address, and

a reading mechanism that distinguishes the updated version from the original data at least in part by the relative time from said indication of the later writing of the updated version.