WO2008137755A1 - Wear leveling - Google Patents

Wear leveling Download PDF

Info

Publication number
WO2008137755A1
WO2008137755A1 PCT/US2008/062520 US2008062520W WO2008137755A1 WO 2008137755 A1 WO2008137755 A1 WO 2008137755A1 US 2008062520 W US2008062520 W US 2008062520W WO 2008137755 A1 WO2008137755 A1 WO 2008137755A1
Authority
WO
WIPO (PCT)
Prior art keywords
memory location
memory
allocated
leveling
identifying
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US2008/062520
Other languages
French (fr)
Inventor
Russell Hobson
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Atmel Corp
Original Assignee
Atmel Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Atmel Corp filed Critical Atmel Corp
Priority to DE112008001126.5T priority Critical patent/DE112008001126B4/en
Priority to CN200880014680.5A priority patent/CN101675479B/en
Publication of WO2008137755A1 publication Critical patent/WO2008137755A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C16/00Erasable programmable read-only memories
    • G11C16/02Erasable programmable read-only memories electrically programmable
    • G11C16/06Auxiliary circuits, e.g. for writing into memory
    • G11C16/34Determination of programming status, e.g. threshold voltage, overprogramming or underprogramming, retention
    • G11C16/349Arrangements for evaluating degradation, retention or wearout, e.g. by counting erase cycles

Definitions

  • This disclosure relates to wear leveling.
  • the memory cells of a memory device can eventually wear out from repeated programming and erasure.
  • addressable sections of the memory device can be programmed and/or erased at similar rates by wear leveling.
  • Wear leveling ensures that the memory cells of the memory device wear evenly, e.g., programming and/or erasing of memory cells occurs at a similar rate for all the memory cells over the life of the memory device.
  • Wear leveling typically includes the swapping of data stored in memory locations that are infrequently changed with data stored in frequently changing memory locations, or the moving of data stored in memory blocks that are infrequently changed to unallocated memory blocks. The even wearing of the memory cells can thus prevent premature failure of the memory device.
  • a reference memory location can be designated in a memory device, and a memory location can be identified in response to storing data in the memory device. If the identified memory location is associated with the reference memory location, then an allocated memory location relative to the reference memory location is identified, and the allocated memory location is leveled.
  • a determination can be made as to whether to level an allocated memory location that is relative to a designated reference memory location. If the allocated memory location is determined to be leveled, then an unallocated memory location relative to the allocated memory location can be allocated to store data stored in the allocated memory location, and the allocated memory location can be deallocated. The designated reference memory location can be redesignated relative to the deallocated memory.
  • a first allocated memory location relative to a reference memory location can be identified, and an unallocated memory location relative to the allocated memory location can also be identified.
  • the unallocated memory location can be allocated to store data stored in the first allocated memory location, and the first allocated memory location can be deallocated.
  • a memory location relative to the deallocated memory location can be identified, and the identified memory location can be designated as the reference memory location.
  • a memory device can comprise a memory array comprising memory cells defining memory blocks and a memory driver configured to define a leveling locator and a free block locator.
  • the memory driver can also be configured to reference respective memory blocks indicated by the leveling locator and free block locator, and determine if the free block locator is associated with the leveling locator. Upon a positive determination, the memory driver can identify a utilized memory block respective to the leveling locator, and level the utilized memory block.
  • Fig. 1 is a block diagram of an example computer device.
  • Figs. 2-6 are diagrams of a memory device architecture implementing wear leveling.
  • Fig. 7 is a flow diagram of an example process for wear leveling.
  • Fig. 8 is a flow diagram of an example process for leveling allocated memory.
  • Fig. 9 is a flow diagram of an example process for identifying memory locations for wear leveling.
  • Fig. 10 is a flow diagram of an example process for determining whether to wear level a memory location.
  • Fig. 11 is a flow diagram of another example process of wear leveling.
  • Fig. 1 is a block diagram of an example computer system 100.
  • the system 100 can include a processor 102, a data storage system 104, and a communication system 106. Each of the components 102, 104, and 106 can, for example, be connected by a system bus 108.
  • the processor 102 can process instructions stored in the data storage system 104 or received over the communication system 106.
  • Various different general purpose or special purpose computing system configurations can be implemented as the computer system 100, such as personal computers, server computers, hand-held or laptop devices, portable communication devices, programmable consumer electronics, gaming systems, multimedia systems, etc.
  • the data storage system 104 can, for example, include a volatile memory device, such as a random access memory (RAM), and/or can include a non-volatile memory device, such as a read only memory (ROM) and/or a flash memory. Other volatile and non-volatile memory devices can also be used.
  • the data storage system 104 can, for example, store an operating system and one or more application programs that can be executed by the processor 102, and data associated with the operating system and/or the application programs.
  • the communication system 106 can transmit and receive data over a wired or wireless communication link.
  • the communication subsystem 106 can, for example, include communication devices that support one or more communication protocols, such as the Institute of Electrical and Electronics Engineer (IEEE) 802.x family of protocols (e.g., an Ethernet, a token ring, a wireless local area network, a wireless personal area network, etc.) or some other wired or wireless communication protocol.
  • IEEE Institute of Electrical and Electronics Engineer
  • the data storage system 104 can include a memory driver 110 and a non- volatile storage 112, e.g., a flash memory device.
  • the memory driver 110 can, for example, include software and/or hardware to facilitate control of and access to the non-volatile storage 112.
  • the memory driver 110 can, for example, utilize a logical- to-physical addressing translation for read and write operations.
  • An example memory device architecture utilizing logical-to-physical mapping is shown and described with respect to Fig. 2 below.
  • the memory driver 110 and the non-volatile storage 112 can be integrated on the same substrate 114 or device. In another implementation, the memory driver 110 and the non-volatile storage 112 can be on separate substrates or devices, e.g., the memory driver 110 can be implemented in software and/or hardware within the computer system 100, and separate from the substrate or device implementing the non-volatile storage 112.
  • Fig. 2 is a block diagram of a memory device architecture 200.
  • the memory device architecture 200 can include a logical array 220, a logical-to-physical mapping array 240, and a physical array 260.
  • the mapping array 240 can define a mapping of logical addresses defined by the logical array 220 to the physical addresses defined by the physical array 260.
  • the mapping array 240 can dynamically map m blocks of logical array 220 locations to n blocks of physical array 260 locations.
  • m ⁇ n to ensure that unallocated memory locations for wear leveling are available in the physical array 260.
  • the physical array 260 can be realized by a nonvolatile memory device, e.g., a flash memory device. Other memory devices can also be used, however.
  • the mapping array 240 and the physical array 260 can have associated memory allocation flags 242 and 262, respectively.
  • the memory allocation flags 242 and 262 can indicate, for example, whether the contents of each block associated with each memory allocation flag are valid, e.g., a currently allocated block for stored data, or invalid, e.g., a currently unallocated block that may contain expired data, erased data, or may otherwise be identified as an unallocated block of memory.
  • the logical array 220 is currently defining stored data in at least three addresses, e.g., blocks Lc, Ld and Le.
  • the logical array 220 blocks Lc, Ld and Le are mapped to corresponding physical array 260 blocks PO, P1 and P2, as defined by the mapping array 240.
  • Each of the physical array 260 blocks PO, P1 and P2 store data represented by the abstracted data symbol d.
  • the memory cells of a physical array 260 can eventually wear out from repeated programming and erasure. Additionally, some blocks of the physical array 260 may not be programmed or erased very often, and thus other blocks of the physical array 260 are programmed and erased more often. To extend the overall life the physical array 260, the blocks P0...Pn-1 can be programmed and/or erased at similar rates by wear leveling. Wear leveling ensures that the blocks P0...Pn-1 wear evenly, e.g., programming and/or erasing of the blocks P0...Pn-1 occurs at a similar rate for all the blocks over the lifetime of the physical array 260.
  • a reference memory location can be designated in the physical array 260.
  • the reference memory location can be designated by, for example, a leveling pointer LP that stores the location of the reference memory location, e.g. the address PO of block PO.
  • Another identified memory location e.g., a memory location currently eligible to receive data for storage, such as an unallocated memory location, can be designated in the physical array 260.
  • the identified memory location can, for example, be designated by a free block pointer FBP that stores the address of the identified memory location.
  • the leveling pointer LP and the free block pointer FBP can reference the same memory location, e.g., block PO in the physical array 260.
  • the data are stored beginning at the memory block referenced by the free block pointer FBP.
  • the free block pointer FBP is incremented to identify the next memory location eligible to receive data for storage. For example, as shown in Fig. 2, after an initialization, data d are written to logical addresses defined by address blocks Lc, Ld and Le in the logical array 220.
  • a first write operation stores data at the logical address Le; because the free block pointer FBP references memory block PO in the physical array for the first write operation, the data are stored in the physical array at block PO, as indicated by the label "Block P0:d.”
  • the mapping table 240 stores the address of block PO in a table entry associated with the logical address Le, and the corresponding memory allocation flags 242 and 262 for the mapping table 240 and the physical array 260 are set to 1 , respectively.
  • the free block pointer FBP can be incremented to the next available memory block in the physical array 260, e.g., block P1.
  • the free block pointer FBP references the memory block P3 in the physical array 260.
  • the process continues for each subsequent write operation, and the free block pointer FBP is incremented through the physical array 260.
  • the free block pointer FBP traverses the entire physical array 260, at which time the free block pointer FBP and the leveling pointer LP reference the same memory location, e.g., block PO in the physical array 260.
  • a process determines if the identified memory location referenced by the free block pointer FBP is associated with the reference memory location referenced by the leveling pointer LP. For example, a process can determine whether the free block pointer FBP is equal to or has incremented past the leveling pointer LP. The determination can be made, for example, by a hardware and/or software logic implementation on a memory device defining the physical array 260, e.g., a control circuit on a flash card, or can be made by an external device, such as a hardware and/or software logic implementation in a memory driver device.
  • the free block pointer has been incremented n times and thus now stores the address of the first block PO in the physical array 260, e.g., after the n th memory block write to the physical array 260 after initialization, the stored address in the free block pointer FBP rolls over to the initial address PO of the physical array 260.
  • the physical array 260 may have unallocated memory blocks.
  • the n th write to the physical array 260 is related to the storing of data in the logical address Le. As shown in Fig.
  • the logical address Le was initially mapped to the memory block PO in the physical array 260; however, the n th write resulted in the logical address Le being mapped to the memory block Pn-1 in the physical array 260. Accordingly, the block PO is deallocated as the data stored in block PO is no longer mapped to a logical address.
  • the deallocated status of the memory block PO is indicated by the corresponding memory allocation flag 262 being set to 0.
  • wear leveling includes identifying an allocated memory location relative to the reference memory location. For example, as shown in Fig. 3, an allocated memory location can be identified by incrementing the leveling pointer LP through the physical array 260 until a memory block having a set allocation flag 260 is identified. Accordingly, the leveling pointer LP in Fig. 3 increments from the reference memory location block PO to block P2, which is the first allocated memory location relative to the reference memory location PO.
  • An unallocated memory location can then be designated to store the data stored in the allocated memory location.
  • another pointer can be instantiated to identify the unallocated memory location. For example, as shown in Fig. 4, a free leveling pointer FLP is instantiated and set equal to the leveling pointer LP. The free leveling pointer FLP is then incremented through the physical array 260 until a memory block having a cleared allocation flag 260 is identified. Accordingly, the free leveling pointer FLP in Fig. 4 increments from the allocated memory location block P2 to block P4, which is the first unallocated memory location relative to the allocated memory location P2.
  • the leveling pointer LP and the free leveling pointer FLP thus identify a memory block to be leveled, e.g., block P2, and the destination, e.g., block P4, of the data presently stored in the block to be leveled.
  • Fig. 5 depicts the mapping after the data stored in the memory block P2 of the physical array 260 is moved to the memory block P4. As a result of the movement of the data, the allocation flag 262 for the memory block P2 is cleared, and the allocation flag 262 for the memory block P4 is set. Additionally, the mapping of the logical block Lc in the mapping array 240 is updated with the address of the logical block P4.
  • the leveling pointer LP can be set to the value of the free leveling pointer FLP, as indicated in Fig. 5, and the free leveling pointer FLP can be deallocated.
  • the leveling pointer LP can also be set to other values.
  • the leveling point LP can be set to the first unallocated memory location that occurs after the memory location referenced by the free leveling pointer FLP; in another implementation, the leveling point LP can remain set to the previously allocated memory location, e.g., block P2 in Fig.
  • the system can continue writing logical and physical blocks as previously described.
  • another leveling operation will occur when the free block pointer FBP is equal to or increments past the leveling pointer LP a second subsequent time, as it is likely that the free block pointer FBP value is not far behind the leveling pointer LP value after a wear leveling operation, and thus may be equal again after a relatively small number of write operations.
  • the free block pointer FBP can be set to the first unallocated memory location after the memory location reference by the leveling pointer LP. Accordingly, a wear leveling may occur when the free block pointer FBP is again equal to or increments past the leveling pointer LP.
  • a process can determine whether a wear leveling operation is to be performed. The determination can be made, for example, by a hardware and/or software logic implementation on a memory device defining the physical array 260, e.g., a control circuit on a flash card, or can be made by an external device, such as a hardware and/or software logic implementation in a memory driver device.
  • the occurrence of a wear leveling condition can invoke a wear leveling operation.
  • a wear leveling condition can occur each time the free block pointer FBP value is equal to or increments past the leveling pointer LP value; in another implementation, a wear leveling condition can occur every m th time the free block pointer FBP value is equal to or increments past the leveling pointer LP value.
  • wear leveling conditions can also be used, such as delaying execution of the wear leveling process until a time during which a user process may not demand processing resources, e.g., during a shut down or start up process; etc.
  • an operating system can mark the memory blocks for swapping and reserve the memory blocks until the leveling process is performed.
  • an operating system can mark the memory blocks for swapping; however, if the memory blocks marked for leveling have been moved or are flagged an unallocated before the leveling process starts, the leveling process can be cancelled.
  • the leveling pointer LP and the free leveling pointer FLP can be combined as a single pointer.
  • Other memory block allocation and deallocation resources can also be used, e.g., another table, such as an allocation table, can be used instead of the pointers and allocation flags described above.
  • Other leveling implementations can also be used. For example, several blocks of memory can be leveled during a leveling operation. As shown in Fig.
  • the free block pointer FBP initially references the last memory block Pn-1 in the physical array 260; after a write operation is performed and data are stored in the physical array 260 memory block Pn-1 , the free block pointer FBP is incremented to the next unallocated memory block, e.g., memory block P1 , and thus increments past the leveling pointer LP, triggering a leveling operation.
  • the leveling operation of Fig. 6, moves a contiguous blocks of memory locations, e.g., the memory locations P2 and P3, to two unallocated memory locations, e.g., memory blocks P4 and P5.
  • non-contiguous blocks of memory can be leveled during a leveling operation, e.g., the next q block of allocated memory can be wear leveled, etc.
  • particular memory blocks can be exempt from wear leveling, e.g., memory block that are identified as failed blocks; or memory block PO, which can be used as a fixed location for system data.
  • Fig. 7 is a flow diagram of an example process 700 of wear leveling.
  • the example process 700 can, for example, be implemented in the non-volatile storage 112, or in a memory driver 110 associated with the non-volatile store 112, or in some other hardware and/or software memory resource.
  • Stage 702 designates a reference memory location in a memory device.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can associate a leveling pointer LP with a memory location to designate the memory location as a reference memory location.
  • Stage 704 identifies a memory location in response to storing data in the memory device.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can associate a free block pointer FBP with a memory location in response to storing data in a memory device.
  • Stage 706 determines if the identified memory location is associated with the reference memory location.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can determine if the free block pointer FBP is equal to or has incremented past the leveling pointer LP.
  • stage 706 determines that the identified memory location is not associated with the reference memory location, the process returns to stage 704. If, however, stage 706 determines that the identified memory location is associated with the reference memory location, then stage 708 identifies an allocated memory location relative to the reference memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can increment the leveling pointer LP to a first allocated memory location, e.g., the memory block P2 of Fig. 3, for example.
  • Stage 708 levels the allocated memory location.
  • the nonvolatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can move the data stored in the first allocated memory location to an unallocated memory location.
  • Fig. 8 is a flow diagram of an example process 800 of leveling allocated memory.
  • the example process 800 can, for example, be implemented in the nonvolatile storage 112, or in a memory driver 110 associated with the non-volatile store 112, or in some other hardware and/or software memory resource.
  • Stage 802 identifies a first unallocated memory location relative to an originally allocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can increment a free leveling pointer FLP from the address of the leveling pointer LP to identify an unallocated memory location.
  • Stage 804 allocates the first unallocated memory location to store the data stored in the originally allocated memory location.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can store data in the unallocated memory location pointed to by the free leveling pointer and set the corresponding allocation flag 262 to 1 , i.e., the data stored in the originally allocated memory location can be moved to an unallocated memory location, and the state of the unallocated memory location can thus be changed to an allocated memory location.
  • Stage 806 deallocates the originally allocated memory location.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can set the corresponding allocation flag of the originally allocated memory location from which data was moved during the leveling process to 0, thus deallocating the originally allocated memory location.
  • Stage 808 designates the reference memory location relative to the deallocated memory location.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can increment the leveling pointer LP from the deallocated memory location to another memory location.
  • Fig. 9 is a flow diagram of an example process 900 for identifying memory locations for wear leveling.
  • the example process 900 can, for example, be implemented in the non-volatile storage 112, or in a memory driver 110 associated with the non-volatile store 112, or in some other hardware and/or software memory resource.
  • Stage 902 associates each memory location with a corresponding memory allocating flag.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can associated each memory location in the physical array 260 with memory allocation flags 262.
  • Stage 904 identifies a memory location associated with a memory allocation flag indicating an allocated memory location.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can identify a memory location with an allocation flag 262 set to 1 and set the leveling point LP equal to the address of the identified memory location.
  • Stage 906 identifies a memory location associated with a memory allocation flag indicating an unallocated memory location.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can identify a memory location with an allocation flag 262 set to 0 and set the free leveling point FLP equal to the address of the identified memory location.
  • Fig. 10 is a flow diagram of an example process 1000 for determining whether to wear level a memory location.
  • the example process 1000 can, for example, be implemented in the non-volatile storage 112, or in a memory driver 110 associated with the non-volatile store 112, or in some other hardware and/or software memory resource.
  • Stage 1002 identifies a memory location in response to storing data in a memory device.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can associate a free block pointer with a memory location in response to storing data in a memory device.
  • Stage 1004 determines whether the identified memory location is associated with a reference memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can determine if the free block pointer FBP is equal to or has incremented past the leveling pointer LP. If stage 1004 determines that the identified memory location is not associated with a reference memory location, the process returns to stage 1002. If, however, stage 1004 determines that the identified memory location is associated with a reference memory location, then stage 1006 determines whether a wear leveling condition is met. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can determine if one or more of the wear leveling conditions described above has occurred.
  • stage 1006 determines that a wear leveling condition is not met, then the process returns to stage 1002. If, however, stage 1006 determines a wear leveling condition is met, then stage 1008 performs a wear leveling process.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can perform one of the wear leveling processed described above.
  • Fig. 11 is a flow diagram of an example process 1100 of wear leveling.
  • the example process 1000 can, for example, be implemented in the non-volatile storage 112, or in a memory driver 110 associated with the non-volatile store 112, or in some other hardware and/or software memory resource.
  • Stage 1102 identifies a first allocated memory location relative to a reference memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can increment a leveling pointer LP from the current address of the leveling pointer LP to identify an allocated memory location. Stage 1104 identifies an unallocated memory location relative to a first allocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can increment a free leveling pointer FLP from the current address of the incremented leveling pointer LP to identify an unallocated memory location.
  • Stage 1106 allocates the unallocated memory location to store data stored in the first allocated memory location.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can store data in the unallocated memory location pointed to by the free leveling pointer and set the corresponding allocation flag 262 to 1.
  • Stage 1108 deallocates the first allocated memory location.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can set the corresponding allocation flag of the memory location from which data was moved during the leveling process to 0.
  • Stage 1110 identifies a memory location relative to the deallocated memory location.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can identify the memory location referenced by the free leveling pointer FLP or a first unallocated memory location incremented from the address of the free leveling pointer FLP.
  • Stage 1112 designates the identified memory location as the reference memory location.
  • the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can set the leveling pointer LP to the address of the memory location identified in stage 1110.
  • Figs. 2 - 11 above describe leveling at the block level.
  • the mapping table 240 and the pointers can be updated when the writing operation increments onto the next physical block.
  • the processes above can be adapted to any memory architecture that includes a sequential addressing scheme.
  • the apparatus, methods, flow diagrams, and structure block diagrams described in this patent document may be implemented in computer processing systems including program code comprising program instructions that are executable by the computer processing system. Other implementations may also be used. Additionally, the flow diagrams and structure block diagrams described in this patent document, which describe particular methods and/or corresponding acts in support of steps and corresponding functions in support of disclosed structural means, may also be utilized to implement corresponding software structures and algorithms, and equivalents thereof.

Landscapes

  • Techniques For Improving Reliability Of Storages (AREA)
  • Memory System (AREA)

Abstract

A reference memory location can be designated in a memory device. A memory location can be designated in response to storing data in the memory device. If the identified memory location is associated with the reference memory location then an allocated memory location can be designated relative to the reference memory location, and the allocated memory location can be leveled.

Description

WEAR LEVELING
BACKGROUND
This disclosure relates to wear leveling.
The memory cells of a memory device, such as a flash memory device, can eventually wear out from repeated programming and erasure. To extend the overall life of a memory device, addressable sections of the memory device, such as blocks or pages, can be programmed and/or erased at similar rates by wear leveling. Wear leveling ensures that the memory cells of the memory device wear evenly, e.g., programming and/or erasing of memory cells occurs at a similar rate for all the memory cells over the life of the memory device. Wear leveling typically includes the swapping of data stored in memory locations that are infrequently changed with data stored in frequently changing memory locations, or the moving of data stored in memory blocks that are infrequently changed to unallocated memory blocks. The even wearing of the memory cells can thus prevent premature failure of the memory device.
SUMMARY
Disclosed herein are systems and methods of wear leveling. In one implementation, a reference memory location can be designated in a memory device, and a memory location can be identified in response to storing data in the memory device. If the identified memory location is associated with the reference memory location, then an allocated memory location relative to the reference memory location is identified, and the allocated memory location is leveled.
In another implementation, a determination can be made as to whether to level an allocated memory location that is relative to a designated reference memory location. If the allocated memory location is determined to be leveled, then an unallocated memory location relative to the allocated memory location can be allocated to store data stored in the allocated memory location, and the allocated memory location can be deallocated. The designated reference memory location can be redesignated relative to the deallocated memory.
In another implementation, a first allocated memory location relative to a reference memory location can be identified, and an unallocated memory location relative to the allocated memory location can also be identified. The unallocated memory location can be allocated to store data stored in the first allocated memory location, and the first allocated memory location can be deallocated. A memory location relative to the deallocated memory location can be identified, and the identified memory location can be designated as the reference memory location. In another implementation, a memory device can comprise a memory array comprising memory cells defining memory blocks and a memory driver configured to define a leveling locator and a free block locator. The memory driver can also be configured to reference respective memory blocks indicated by the leveling locator and free block locator, and determine if the free block locator is associated with the leveling locator. Upon a positive determination, the memory driver can identify a utilized memory block respective to the leveling locator, and level the utilized memory block.
DESCRIPTION OF DRAWINGS
Fig. 1 is a block diagram of an example computer device. Figs. 2-6 are diagrams of a memory device architecture implementing wear leveling.
Fig. 7 is a flow diagram of an example process for wear leveling.
Fig. 8 is a flow diagram of an example process for leveling allocated memory. Fig. 9 is a flow diagram of an example process for identifying memory locations for wear leveling.
Fig. 10 is a flow diagram of an example process for determining whether to wear level a memory location.
Fig. 11 is a flow diagram of another example process of wear leveling.
DETAILED DESCRIPTION
Fig. 1 is a block diagram of an example computer system 100. The system 100 can include a processor 102, a data storage system 104, and a communication system 106. Each of the components 102, 104, and 106 can, for example, be connected by a system bus 108. The processor 102 can process instructions stored in the data storage system 104 or received over the communication system 106. Various different general purpose or special purpose computing system configurations can be implemented as the computer system 100, such as personal computers, server computers, hand-held or laptop devices, portable communication devices, programmable consumer electronics, gaming systems, multimedia systems, etc. The data storage system 104 can, for example, include a volatile memory device, such as a random access memory (RAM), and/or can include a non-volatile memory device, such as a read only memory (ROM) and/or a flash memory. Other volatile and non-volatile memory devices can also be used. The data storage system 104 can, for example, store an operating system and one or more application programs that can be executed by the processor 102, and data associated with the operating system and/or the application programs.
The communication system 106 can transmit and receive data over a wired or wireless communication link. The communication subsystem 106 can, for example, include communication devices that support one or more communication protocols, such as the Institute of Electrical and Electronics Engineer (IEEE) 802.x family of protocols (e.g., an Ethernet, a token ring, a wireless local area network, a wireless personal area network, etc.) or some other wired or wireless communication protocol.
The data storage system 104 can include a memory driver 110 and a non- volatile storage 112, e.g., a flash memory device. The memory driver 110 can, for example, include software and/or hardware to facilitate control of and access to the non-volatile storage 112. The memory driver 110 can, for example, utilize a logical- to-physical addressing translation for read and write operations. An example memory device architecture utilizing logical-to-physical mapping is shown and described with respect to Fig. 2 below.
In one implementation, the memory driver 110 and the non-volatile storage 112 can be integrated on the same substrate 114 or device. In another implementation, the memory driver 110 and the non-volatile storage 112 can be on separate substrates or devices, e.g., the memory driver 110 can be implemented in software and/or hardware within the computer system 100, and separate from the substrate or device implementing the non-volatile storage 112.
Fig. 2 is a block diagram of a memory device architecture 200. The memory device architecture 200 can include a logical array 220, a logical-to-physical mapping array 240, and a physical array 260. The mapping array 240 can define a mapping of logical addresses defined by the logical array 220 to the physical addresses defined by the physical array 260. For example, the mapping array 240 can dynamically map m blocks of logical array 220 locations to n blocks of physical array 260 locations. In some implementation, each logical array 240 block, e.g. Block Lx, where x = 0...m-1 , when allocated, corresponds to a mapped physical array block, e.g., Block Py, where y = 0...n-1. In some implementations, m < n to ensure that unallocated memory locations for wear leveling are available in the physical array 260. Other implementations in which m = n or m > n can also be used, however. In one implementation, the physical array 260 can be realized by a nonvolatile memory device, e.g., a flash memory device. Other memory devices can also be used, however.
The mapping array 240 and the physical array 260 can have associated memory allocation flags 242 and 262, respectively. The memory allocation flags 242 and 262 can indicate, for example, whether the contents of each block associated with each memory allocation flag are valid, e.g., a currently allocated block for stored data, or invalid, e.g., a currently unallocated block that may contain expired data, erased data, or may otherwise be identified as an unallocated block of memory. For example, as shown in Fig. 2, the logical array 220 is currently defining stored data in at least three addresses, e.g., blocks Lc, Ld and Le. The logical array 220 blocks Lc, Ld and Le are mapped to corresponding physical array 260 blocks PO, P1 and P2, as defined by the mapping array 240. Each of the physical array 260 blocks PO, P1 and P2 store data represented by the abstracted data symbol d.
The memory cells of a physical array 260 can eventually wear out from repeated programming and erasure. Additionally, some blocks of the physical array 260 may not be programmed or erased very often, and thus other blocks of the physical array 260 are programmed and erased more often. To extend the overall life the physical array 260, the blocks P0...Pn-1 can be programmed and/or erased at similar rates by wear leveling. Wear leveling ensures that the blocks P0...Pn-1 wear evenly, e.g., programming and/or erasing of the blocks P0...Pn-1 occurs at a similar rate for all the blocks over the lifetime of the physical array 260.
In an implementation, a reference memory location can be designated in the physical array 260. The reference memory location can be designated by, for example, a leveling pointer LP that stores the location of the reference memory location, e.g. the address PO of block PO. Another identified memory location, e.g., a memory location currently eligible to receive data for storage, such as an unallocated memory location, can be designated in the physical array 260. The identified memory location can, for example, be designated by a free block pointer FBP that stores the address of the identified memory location. Upon a memory device initialization, e.g., a clearing of an entire flash memory, for example, the leveling pointer LP and the free block pointer FBP can reference the same memory location, e.g., block PO in the physical array 260. When data are stored in the physical array 260, the data are stored beginning at the memory block referenced by the free block pointer FBP. After the data are stored, the free block pointer FBP is incremented to identify the next memory location eligible to receive data for storage. For example, as shown in Fig. 2, after an initialization, data d are written to logical addresses defined by address blocks Lc, Ld and Le in the logical array 220. A first write operation stores data at the logical address Le; because the free block pointer FBP references memory block PO in the physical array for the first write operation, the data are stored in the physical array at block PO, as indicated by the label "Block P0:d." The mapping table 240 stores the address of block PO in a table entry associated with the logical address Le, and the corresponding memory allocation flags 242 and 262 for the mapping table 240 and the physical array 260 are set to 1 , respectively. The free block pointer FBP can be incremented to the next available memory block in the physical array 260, e.g., block P1.
After several more write operations, e.g., writes to logical addresses Lc and Ld, respectively, the free block pointer FBP references the memory block P3 in the physical array 260. The process continues for each subsequent write operation, and the free block pointer FBP is incremented through the physical array 260. Eventually the free block pointer FBP traverses the entire physical array 260, at which time the free block pointer FBP and the leveling pointer LP reference the same memory location, e.g., block PO in the physical array 260.
In some implementations, after each increment of the free block pointer FBP, a process determines if the identified memory location referenced by the free block pointer FBP is associated with the reference memory location referenced by the leveling pointer LP. For example, a process can determine whether the free block pointer FBP is equal to or has incremented past the leveling pointer LP. The determination can be made, for example, by a hardware and/or software logic implementation on a memory device defining the physical array 260, e.g., a control circuit on a flash card, or can be made by an external device, such as a hardware and/or software logic implementation in a memory driver device.
For example, as shown in Fig. 3, the free block pointer has been incremented n times and thus now stores the address of the first block PO in the physical array 260, e.g., after the nth memory block write to the physical array 260 after initialization, the stored address in the free block pointer FBP rolls over to the initial address PO of the physical array 260. Note, however, that after the nth memory block write to the physical array 260, the physical array 260 may have unallocated memory blocks. For example, the nth write to the physical array 260 is related to the storing of data in the logical address Le. As shown in Fig. 2, the logical address Le was initially mapped to the memory block PO in the physical array 260; however, the nth write resulted in the logical address Le being mapped to the memory block Pn-1 in the physical array 260. Accordingly, the block PO is deallocated as the data stored in block PO is no longer mapped to a logical address. The deallocated status of the memory block PO is indicated by the corresponding memory allocation flag 262 being set to 0.
In some implementations, if the identified memory location referenced by the free block pointer FBP is associated with the reference memory location referenced by the leveling pointer LP, a wear leveling operation can be performed. In one implementation, wear leveling includes identifying an allocated memory location relative to the reference memory location. For example, as shown in Fig. 3, an allocated memory location can be identified by incrementing the leveling pointer LP through the physical array 260 until a memory block having a set allocation flag 260 is identified. Accordingly, the leveling pointer LP in Fig. 3 increments from the reference memory location block PO to block P2, which is the first allocated memory location relative to the reference memory location PO.
An unallocated memory location can then be designated to store the data stored in the allocated memory location. In some implementations, another pointer can be instantiated to identify the unallocated memory location. For example, as shown in Fig. 4, a free leveling pointer FLP is instantiated and set equal to the leveling pointer LP. The free leveling pointer FLP is then incremented through the physical array 260 until a memory block having a cleared allocation flag 260 is identified. Accordingly, the free leveling pointer FLP in Fig. 4 increments from the allocated memory location block P2 to block P4, which is the first unallocated memory location relative to the allocated memory location P2.
The leveling pointer LP and the free leveling pointer FLP thus identify a memory block to be leveled, e.g., block P2, and the destination, e.g., block P4, of the data presently stored in the block to be leveled. Fig. 5 depicts the mapping after the data stored in the memory block P2 of the physical array 260 is moved to the memory block P4. As a result of the movement of the data, the allocation flag 262 for the memory block P2 is cleared, and the allocation flag 262 for the memory block P4 is set. Additionally, the mapping of the logical block Lc in the mapping array 240 is updated with the address of the logical block P4. In one implementation, after the data stored in the allocated memory location, e.g., memory block P2, is moved to the unallocated memory location, e.g., memory block P4, the leveling pointer LP can be set to the value of the free leveling pointer FLP, as indicated in Fig. 5, and the free leveling pointer FLP can be deallocated. The leveling pointer LP, however, can also be set to other values. For example, in one implementation, the leveling point LP can be set to the first unallocated memory location that occurs after the memory location referenced by the free leveling pointer FLP; in another implementation, the leveling point LP can remain set to the previously allocated memory location, e.g., block P2 in Fig. 5; or can be set to some other value. Once the wear leveling process is complete, the system can continue writing logical and physical blocks as previously described. In some implementations, another leveling operation will occur when the free block pointer FBP is equal to or increments past the leveling pointer LP a second subsequent time, as it is likely that the free block pointer FBP value is not far behind the leveling pointer LP value after a wear leveling operation, and thus may be equal again after a relatively small number of write operations.
In some implementations, the free block pointer FBP can be set to the first unallocated memory location after the memory location reference by the leveling pointer LP. Accordingly, a wear leveling may occur when the free block pointer FBP is again equal to or increments past the leveling pointer LP.
In some implementations, each time the value of the free block pointer FBP is equal to or increments past the value of the leveling pointer LP, a process can determine whether a wear leveling operation is to be performed. The determination can be made, for example, by a hardware and/or software logic implementation on a memory device defining the physical array 260, e.g., a control circuit on a flash card, or can be made by an external device, such as a hardware and/or software logic implementation in a memory driver device. In one implementation, the occurrence of a wear leveling condition can invoke a wear leveling operation. For example, in one implementation, a wear leveling condition can occur each time the free block pointer FBP value is equal to or increments past the leveling pointer LP value; in another implementation, a wear leveling condition can occur every mth time the free block pointer FBP value is equal to or increments past the leveling pointer LP value.
Other wear leveling conditions can also be used, such as delaying execution of the wear leveling process until a time during which a user process may not demand processing resources, e.g., during a shut down or start up process; etc. In one implementation, if the leveling pointer LP and the free block pointer FBP are equal, an operating system can mark the memory blocks for swapping and reserve the memory blocks until the leveling process is performed. In another implementation, if the leveling pointer LP and the free block pointer FBP are equal, an operating system can mark the memory blocks for swapping; however, if the memory blocks marked for leveling have been moved or are flagged an unallocated before the leveling process starts, the leveling process can be cancelled.
In some implementations, the leveling pointer LP and the free leveling pointer FLP can be combined as a single pointer. Other memory block allocation and deallocation resources can also be used, e.g., another table, such as an allocation table, can be used instead of the pointers and allocation flags described above. Other leveling implementations can also be used. For example, several blocks of memory can be leveled during a leveling operation. As shown in Fig. 6, the free block pointer FBP initially references the last memory block Pn-1 in the physical array 260; after a write operation is performed and data are stored in the physical array 260 memory block Pn-1 , the free block pointer FBP is incremented to the next unallocated memory block, e.g., memory block P1 , and thus increments past the leveling pointer LP, triggering a leveling operation. The leveling operation of Fig. 6, however, moves a contiguous blocks of memory locations, e.g., the memory locations P2 and P3, to two unallocated memory locations, e.g., memory blocks P4 and P5.
In some implementations, several non-contiguous blocks of memory can be leveled during a leveling operation, e.g., the next q block of allocated memory can be wear leveled, etc. In some implementations, particular memory blocks can be exempt from wear leveling, e.g., memory block that are identified as failed blocks; or memory block PO, which can be used as a fixed location for system data.
Fig. 7 is a flow diagram of an example process 700 of wear leveling. The example process 700 can, for example, be implemented in the non-volatile storage 112, or in a memory driver 110 associated with the non-volatile store 112, or in some other hardware and/or software memory resource.
Stage 702 designates a reference memory location in a memory device. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can associate a leveling pointer LP with a memory location to designate the memory location as a reference memory location. Stage 704 identifies a memory location in response to storing data in the memory device. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can associate a free block pointer FBP with a memory location in response to storing data in a memory device.
Stage 706 determines if the identified memory location is associated with the reference memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can determine if the free block pointer FBP is equal to or has incremented past the leveling pointer LP.
If stage 706 determines that the identified memory location is not associated with the reference memory location, the process returns to stage 704. If, however, stage 706 determines that the identified memory location is associated with the reference memory location, then stage 708 identifies an allocated memory location relative to the reference memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can increment the leveling pointer LP to a first allocated memory location, e.g., the memory block P2 of Fig. 3, for example.
Stage 708 levels the allocated memory location. For example, the nonvolatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can move the data stored in the first allocated memory location to an unallocated memory location.
Fig. 8 is a flow diagram of an example process 800 of leveling allocated memory. The example process 800 can, for example, be implemented in the nonvolatile storage 112, or in a memory driver 110 associated with the non-volatile store 112, or in some other hardware and/or software memory resource.
Stage 802 identifies a first unallocated memory location relative to an originally allocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can increment a free leveling pointer FLP from the address of the leveling pointer LP to identify an unallocated memory location.
Stage 804 allocates the first unallocated memory location to store the data stored in the originally allocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can store data in the unallocated memory location pointed to by the free leveling pointer and set the corresponding allocation flag 262 to 1 , i.e., the data stored in the originally allocated memory location can be moved to an unallocated memory location, and the state of the unallocated memory location can thus be changed to an allocated memory location.
Stage 806 deallocates the originally allocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can set the corresponding allocation flag of the originally allocated memory location from which data was moved during the leveling process to 0, thus deallocating the originally allocated memory location.
Stage 808 designates the reference memory location relative to the deallocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can increment the leveling pointer LP from the deallocated memory location to another memory location. Fig. 9 is a flow diagram of an example process 900 for identifying memory locations for wear leveling. The example process 900 can, for example, be implemented in the non-volatile storage 112, or in a memory driver 110 associated with the non-volatile store 112, or in some other hardware and/or software memory resource.
Stage 902 associates each memory location with a corresponding memory allocating flag. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can associated each memory location in the physical array 260 with memory allocation flags 262. Stage 904 identifies a memory location associated with a memory allocation flag indicating an allocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can identify a memory location with an allocation flag 262 set to 1 and set the leveling point LP equal to the address of the identified memory location. Stage 906 identifies a memory location associated with a memory allocation flag indicating an unallocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can identify a memory location with an allocation flag 262 set to 0 and set the free leveling point FLP equal to the address of the identified memory location. Fig. 10 is a flow diagram of an example process 1000 for determining whether to wear level a memory location. The example process 1000 can, for example, be implemented in the non-volatile storage 112, or in a memory driver 110 associated with the non-volatile store 112, or in some other hardware and/or software memory resource. Stage 1002 identifies a memory location in response to storing data in a memory device. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can associate a free block pointer with a memory location in response to storing data in a memory device.
Stage 1004 determines whether the identified memory location is associated with a reference memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can determine if the free block pointer FBP is equal to or has incremented past the leveling pointer LP. If stage 1004 determines that the identified memory location is not associated with a reference memory location, the process returns to stage 1002. If, however, stage 1004 determines that the identified memory location is associated with a reference memory location, then stage 1006 determines whether a wear leveling condition is met. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can determine if one or more of the wear leveling conditions described above has occurred.
If stage 1006 determines that a wear leveling condition is not met, then the process returns to stage 1002. If, however, stage 1006 determines a wear leveling condition is met, then stage 1008 performs a wear leveling process. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can perform one of the wear leveling processed described above.
Fig. 11 is a flow diagram of an example process 1100 of wear leveling. The example process 1000 can, for example, be implemented in the non-volatile storage 112, or in a memory driver 110 associated with the non-volatile store 112, or in some other hardware and/or software memory resource.
Stage 1102 identifies a first allocated memory location relative to a reference memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can increment a leveling pointer LP from the current address of the leveling pointer LP to identify an allocated memory location. Stage 1104 identifies an unallocated memory location relative to a first allocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can increment a free leveling pointer FLP from the current address of the incremented leveling pointer LP to identify an unallocated memory location.
Stage 1106 allocates the unallocated memory location to store data stored in the first allocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can store data in the unallocated memory location pointed to by the free leveling pointer and set the corresponding allocation flag 262 to 1.
Stage 1108 deallocates the first allocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can set the corresponding allocation flag of the memory location from which data was moved during the leveling process to 0.
Stage 1110 identifies a memory location relative to the deallocated memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can identify the memory location referenced by the free leveling pointer FLP or a first unallocated memory location incremented from the address of the free leveling pointer FLP.
Stage 1112 designates the identified memory location as the reference memory location. For example, the non-volatile storage 112, or a memory driver 110 associated with the non-volatile store 112 can set the leveling pointer LP to the address of the memory location identified in stage 1110.
Figs. 2 - 11 above describe leveling at the block level. For memory devices in which writing operations are made on a page basis, the mapping table 240 and the pointers can be updated when the writing operation increments onto the next physical block. Similarly, the processes above can be adapted to any memory architecture that includes a sequential addressing scheme.
The apparatus, methods, flow diagrams, and structure block diagrams described in this patent document may be implemented in computer processing systems including program code comprising program instructions that are executable by the computer processing system. Other implementations may also be used. Additionally, the flow diagrams and structure block diagrams described in this patent document, which describe particular methods and/or corresponding acts in support of steps and corresponding functions in support of disclosed structural means, may also be utilized to implement corresponding software structures and algorithms, and equivalents thereof.
This written description sets forth the best mode of the invention and provides examples to describe the invention and to enable a person of ordinary skill in the art to make and use the invention. This written description does not limit the invention to the precise terms set forth. Thus, while the invention has been described in detail with reference to the examples set forth above, those of ordinary skill in the art may effect alterations, modifications and variations to the examples without departing from the scope of the invention.

Claims

WHAT IS CLAIMED IS:
1. A method, comprising: designating a reference memory location in a memory device; identifying a memory location in response to storing data in the memory device; determining if the identified memory location is associated with the reference memory location; if the identified memory location is associated with the reference memory location, then: identifying an allocated memory location relative to the reference memory location; and leveling the allocated memory location.
2. The method of claim 1 , wherein leveling the allocated memory location comprises: identifying a first unallocated memory location relative to the allocated memory location; allocating the first unallocated memory location to store data stored in the allocated memory location; and deallocating the allocated memory location.
3. The method of claim 2, comprising: designating the reference memory location relative to the deallocated memory location.
4. The method of claim 1 , wherein identifying a memory location comprises: sequentially identifying memory locations in response to storing data in the memory device.
5. The method of claim 1 , wherein determining if an identified memory location is associated with the reference memory location comprises: determining if the identified memory location is equal to the reference memory location.
6. The method of claim 1 , wherein the memory device comprises a flash memory device.
7. The method of claim 3, wherein designating the reference memory location relative to the deallocated memory location comprises: designating a next sequential allocated memory location relative to the location of the deallocated memory location.
8. The method of claim 3, wherein designating the reference memory location relative to the deallocated memory location comprises: designating a next sequential unallocated memory location relative to the location of the deallocated memory location.
9. The method of claim 2, comprising: associating each memory location with a corresponding memory allocation flag.
10. The method of claim 9, wherein identifying an allocated memory location relative to the reference memory location comprises: identifying a memory location associated with a memory allocation flag indicating an allocated memory location.
11. The method of claim 9, wherein identifying a first unallocated memory location relative to the allocated memory location comprises: identifying a memory location with a memory allocation flag indicating an unallocated memory location.
12. The method of claim 1 , wherein identifying an allocated memory location relative to the reference memory location comprises: incrementing a pointer associated with the reference memory location.
13. The method of claim 12, wherein leveling the allocated memory location comprises: leveling the allocation memory location based on the pointer.
14. A method , comprising: determining whether to level an allocated memory location, wherein the allocated memory location is relative to a designated reference memory location; if the allocated memory location is determined to be leveled, then: allocating an unallocated memory location relative to the allocated memory location to store data stored in the allocated memory location; deallocating the allocated memory location; and redesignating the designated reference memory location relative to the deallocated memory location.
15. The method of claim 14, wherein determining whether to level an allocated memory location comprises: identifying memory locations in response to storing data in the memory device; and determining if an identified memory location is associated with the designated reference memory location.
16. The method of claim 15, comprising: incrementing a counter associated with the designated reference memory location; and leveling the allocation memory location based on the counter.
17. The method of claim 15, wherein determining if an identified memory location is associated with the designated reference memory location comprises: determining if the identified memory location is equal to the designated reference memory location.
18. A method, comprising: identifying a first allocated memory location relative to a reference memory location; identifying an unallocated memory location relative to a first allocated memory location; allocating the unallocated memory location to store data stored in the first allocated memory location; deallocating the first allocated memory location; identifying a memory location relative to the deallocated memory location; and designating the identified memory location as the reference memory location.
19. The method of claim 18, wherein identifying a memory location relative to the deallocated memory location comprises: identifying an unallocated memory location relative to the deallocated memory location.
20. The method of claim 18, wherein identifying a memory location relative to the deallocated memory location comprises: identifying an allocated memory location relative to the deallocated memory location.
21. A memory device, comprising: a memory array comprising memory cells defining memory blocks; and a memory control circuit configured to: define a leveling locator and a free block locator; reference respective memory blocks indicated by the leveling locator and free block locator; determine if the free block locator is associated with the leveling locator; and upon a positive determination, identify a utilized memory block respective to the leveling locator; and level the utilized memory block.
22. The memory device of claim 21 , wherein the memory control circuit is configured to: define a free leveling locator respective to the utilized memory block; and allocate the memory block associated with the free leveling locator to store data stored in the utilized memory block; and deallocate the memory block associated with the utilized memory block.
23. The memory device of claim 22, wherein the memory control circuit is configured to: redefine the leveling locator with respect to the deallocated memory block.
24. The memory device of claim 21 , wherein each memory cell is associated with a corresponding memory allocation flag.
25. The memory device of claim 21 , wherein the memory device comprises a flash memory.
26. The memory device of claim 21 , wherein the memory array and the memory driver are on the same substrate.
27. A system, comprising: means for identifying a first allocated memory location relative to a reference memory location; means for identifying an unallocated memory location relative to a first allocated memory location; means for allocating the unallocated memory location to store data stored in the first allocated memory location; means for deallocating the first allocated memory location; and means for identifying a memory location relative to the deallocated memory location; and means for designating the identified memory location as the reference memory location.
PCT/US2008/062520 2007-05-03 2008-05-02 Wear leveling Ceased WO2008137755A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
DE112008001126.5T DE112008001126B4 (en) 2007-05-03 2008-05-02 wear compensation
CN200880014680.5A CN101675479B (en) 2007-05-03 2008-05-02 Storage Wear Leveling

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US11/743,974 US7689762B2 (en) 2007-05-03 2007-05-03 Storage device wear leveling
US11/743,974 2007-05-03

Publications (1)

Publication Number Publication Date
WO2008137755A1 true WO2008137755A1 (en) 2008-11-13

Family

ID=39940390

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2008/062520 Ceased WO2008137755A1 (en) 2007-05-03 2008-05-02 Wear leveling

Country Status (5)

Country Link
US (1) US7689762B2 (en)
CN (1) CN101675479B (en)
DE (1) DE112008001126B4 (en)
TW (1) TWI442228B (en)
WO (1) WO2008137755A1 (en)

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100185806A1 (en) * 2009-01-16 2010-07-22 Arvind Pruthi Caching systems and methods using a solid state disk
US8671258B2 (en) 2009-03-27 2014-03-11 Lsi Corporation Storage system logical block address de-allocation management
US8230159B2 (en) * 2009-03-27 2012-07-24 Lsi Corporation System, method, and computer program product for sending logical block address de-allocation status information
US20100250830A1 (en) * 2009-03-27 2010-09-30 Ross John Stenfort System, method, and computer program product for hardening data stored on a solid state disk
US8090905B2 (en) * 2009-03-27 2012-01-03 Sandforce, Inc. System, method, and computer program product for converting logical block address de-allocation information in a first format to a second format
US8051241B2 (en) * 2009-05-07 2011-11-01 Seagate Technology Llc Wear leveling technique for storage devices
US20100318719A1 (en) * 2009-06-12 2010-12-16 Micron Technology, Inc. Methods, memory controllers and devices for wear leveling a memory
US9792074B2 (en) * 2009-07-06 2017-10-17 Seagate Technology Llc System, method, and computer program product for interfacing one or more storage devices with a plurality of bridge chips
US8539139B1 (en) 2010-12-17 2013-09-17 Teradota Us, Inc. Managing device wearout using I/O metering
US9477590B2 (en) * 2011-09-16 2016-10-25 Apple Inc. Weave sequence counter for non-volatile memory systems
US20130097403A1 (en) 2011-10-18 2013-04-18 Rambus Inc. Address Mapping in Memory Systems
US9116792B2 (en) * 2012-05-18 2015-08-25 Silicon Motion, Inc. Data storage device and method for flash block management
CN102981971B (en) * 2012-12-25 2016-06-08 重庆大学 A kind of phase transition storage loss equalizing method of quick response
CN103902463A (en) * 2012-12-31 2014-07-02 杨威锋 High-speed physical address assignment technology of memory device based on flash memories
CN103218177A (en) * 2013-04-19 2013-07-24 杭州电子科技大学 NAND Flash storage system with function of data balanced processing
WO2014196000A1 (en) * 2013-06-03 2014-12-11 株式会社日立製作所 Storage device and storage device control method
US10802936B2 (en) 2015-09-14 2020-10-13 Hewlett Packard Enterprise Development Lp Memory location remapping and wear-levelling
KR20170045406A (en) * 2015-10-16 2017-04-27 에스케이하이닉스 주식회사 Data storage device and operating method thereof
US10157141B2 (en) 2016-03-09 2018-12-18 Toshiba Memory Corporation Memory system and method of controlling nonvolatile memory
US20170286311A1 (en) * 2016-04-01 2017-10-05 Dale J. Juenemann Repetitive address indirection in a memory
US10884889B2 (en) 2018-06-22 2021-01-05 Seagate Technology Llc Allocating part of a raid stripe to repair a second raid stripe
US11119946B2 (en) 2019-05-16 2021-09-14 Micron Technology, Inc. Codeword rotation for zone grouping of media codewords

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030033510A1 (en) * 2001-08-08 2003-02-13 David Dice Methods and apparatus for controlling speculative execution of instructions based on a multiaccess memory condition
US20030229766A1 (en) * 2002-06-06 2003-12-11 David Dice Methods and apparatus for performing a memory management technique
US20050047229A1 (en) * 2002-12-18 2005-03-03 Benoit Nadeau-Dostie Method and circuit for collecting memory failure information

Family Cites Families (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5537552A (en) * 1990-11-27 1996-07-16 Canon Kabushiki Kaisha Apparatus for selectively comparing pointers to detect full or empty status of a circular buffer area in an input/output (I/O) buffer
US6230233B1 (en) 1991-09-13 2001-05-08 Sandisk Corporation Wear leveling techniques for flash EEPROM systems
US5404485A (en) 1993-03-08 1995-04-04 M-Systems Flash Disk Pioneers Ltd. Flash file system
US5930815A (en) * 1995-07-31 1999-07-27 Lexar Media, Inc. Moving sequential sectors within a block of information in a flash memory mass storage architecture
GB2317720A (en) 1996-09-30 1998-04-01 Nokia Mobile Phones Ltd Managing Flash memory
US5956473A (en) 1996-11-25 1999-09-21 Macronix International Co., Ltd. Method and system for managing a flash memory mass storage system
US6000006A (en) 1997-08-25 1999-12-07 Bit Microsystems, Inc. Unified re-map and cache-index table with dual write-counters for wear-leveling of non-volatile flash RAM mass storage
US5937425A (en) 1997-10-16 1999-08-10 M-Systems Flash Disk Pioneers Ltd. Flash file system optimized for page-mode flash technologies
KR100297986B1 (en) 1998-03-13 2001-10-25 김영환 Wear levelling system of flash memory cell array and wear levelling method thereof
GB9903490D0 (en) 1999-02-17 1999-04-07 Memory Corp Plc Memory system
US6427186B1 (en) 1999-03-30 2002-07-30 Frank (Fong-Long) Lin Memory, interface system and method for mapping logical block numbers to physical sector numbers in a flash memory, using a master index table and a table of physical sector numbers
US6732221B2 (en) 2001-06-01 2004-05-04 M-Systems Flash Disk Pioneers Ltd Wear leveling of static areas in flash memory
US20030058681A1 (en) 2001-09-27 2003-03-27 Intel Corporation Mechanism for efficient wearout counters in destructive readout memory
GB0123412D0 (en) * 2001-09-28 2001-11-21 Memquest Ltd Memory system sectors
US20030163633A1 (en) * 2002-02-27 2003-08-28 Aasheim Jered Donald System and method for achieving uniform wear levels in a flash memory device
KR100453053B1 (en) 2002-06-10 2004-10-15 삼성전자주식회사 Flash memory file system
KR100484147B1 (en) 2002-07-26 2005-04-18 삼성전자주식회사 Flash memory management method
US6973531B1 (en) 2002-10-28 2005-12-06 Sandisk Corporation Tracking the most frequently erased blocks in non-volatile memory systems
KR101122511B1 (en) * 2002-10-28 2012-03-15 쌘디스크 코포레이션 Automated wear leveling in non-volatile storage systems
US7035967B2 (en) 2002-10-28 2006-04-25 Sandisk Corporation Maintaining an average erase count in a non-volatile storage system
US6985992B1 (en) 2002-10-28 2006-01-10 Sandisk Corporation Wear-leveling in non-volatile storage systems
US7096313B1 (en) 2002-10-28 2006-08-22 Sandisk Corporation Tracking the least frequently erased blocks in non-volatile memory systems
US20050055495A1 (en) * 2003-09-05 2005-03-10 Nokia Corporation Memory wear leveling
US7139863B1 (en) 2003-09-26 2006-11-21 Storage Technology Corporation Method and system for improving usable life of memory devices using vector processing
US7441067B2 (en) * 2004-11-15 2008-10-21 Sandisk Corporation Cyclic flash memory wear leveling
US7315917B2 (en) * 2005-01-20 2008-01-01 Sandisk Corporation Scheduling of housekeeping operations in flash memory systems
US20060161724A1 (en) * 2005-01-20 2006-07-20 Bennett Alan D Scheduling of housekeeping operations in flash memory systems
US20060256623A1 (en) 2005-05-12 2006-11-16 Micron Technology, Inc. Partial string erase scheme in a flash memory device
JP4688584B2 (en) 2005-06-21 2011-05-25 株式会社日立製作所 Storage device
US7853749B2 (en) * 2005-09-01 2010-12-14 Cypress Semiconductor Corporation Flash drive fast wear leveling
JP2007133683A (en) * 2005-11-10 2007-05-31 Sony Corp Memory system
US20070150644A1 (en) * 2005-12-28 2007-06-28 Yosi Pinto System for writing non-volatile memories for increased endurance
US20070208904A1 (en) * 2006-03-03 2007-09-06 Wu-Han Hsieh Wear leveling method and apparatus for nonvolatile memory

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030033510A1 (en) * 2001-08-08 2003-02-13 David Dice Methods and apparatus for controlling speculative execution of instructions based on a multiaccess memory condition
US20030229766A1 (en) * 2002-06-06 2003-12-11 David Dice Methods and apparatus for performing a memory management technique
US20050047229A1 (en) * 2002-12-18 2005-03-03 Benoit Nadeau-Dostie Method and circuit for collecting memory failure information

Also Published As

Publication number Publication date
TW200900927A (en) 2009-01-01
CN101675479B (en) 2013-06-12
DE112008001126T5 (en) 2010-03-25
US7689762B2 (en) 2010-03-30
TWI442228B (en) 2014-06-21
DE112008001126B4 (en) 2018-07-12
US20080276035A1 (en) 2008-11-06
CN101675479A (en) 2010-03-17

Similar Documents

Publication Publication Date Title
US7689762B2 (en) Storage device wear leveling
CN109725847B (en) Memory system and control method
CN110781096B (en) Apparatus and method for performing garbage collection by predicting demand time
KR101788332B1 (en) Mount-time unmapping of unused logical addresses in non-volatile memory systems
US7594067B2 (en) Enhanced data access in a storage device
US8769189B2 (en) Method and apparatus for byte-access in block-based flash memory
US8244959B2 (en) Software adapted wear leveling
US8453021B2 (en) Wear leveling in solid-state device
CN101208669B (en) Technique to write to a non-volatile memory
US20080189490A1 (en) Memory mapping
US7287117B2 (en) Flash memory and mapping control apparatus and method for flash memory
US10073771B2 (en) Data storage method and system thereof
US20150277786A1 (en) Method, device, and program for managing a flash memory for mass storage
JP6908789B2 (en) Multi-level addressing
CN107301015B (en) System and method for reducing stress on memory devices
CN101901190B (en) Electronic device, apparatus and method for managing memory in the electronic device
US7426605B2 (en) Method and apparatus for optimizing flash device erase distribution
CN111367823B (en) Method and device for writing effective data
US8510533B2 (en) Method of managing data on a non-volatile memory
EP3948550B1 (en) An apparatus, method and computer program for managing memory page updates within non-volatile memory
US20150220432A1 (en) Method and apparatus for managing at least one non-volatile memory
US20110055459A1 (en) Method for managing a plurality of blocks of a flash memory, and associated memory device and controller thereof

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 200880014680.5

Country of ref document: CN

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

Ref document number: 08769280

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 1120080011265

Country of ref document: DE

RET De translation (de og part 6b)

Ref document number: 112008001126

Country of ref document: DE

Date of ref document: 20100325

Kind code of ref document: P

122 Ep: pct application non-entry in european phase

Ref document number: 08769280

Country of ref document: EP

Kind code of ref document: A1