US20060018166A1 - Method and managing bad memory blocks in a nonvolatile memory device, and nonvolatile-memory device implementing the management method - Google Patents

Method and managing bad memory blocks in a nonvolatile memory device, and nonvolatile-memory device implementing the management method Download PDF

Info

Publication number
US20060018166A1
US20060018166A1 US11/152,675 US15267505A US2006018166A1 US 20060018166 A1 US20060018166 A1 US 20060018166A1 US 15267505 A US15267505 A US 15267505A US 2006018166 A1 US2006018166 A1 US 2006018166A1
Authority
US
United States
Prior art keywords
bad
memory blocks
memory block
address
spare
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.)
Abandoned
Application number
US11/152,675
Inventor
Massimo Iaculo
Nicola Guida
Andrea Ruggiero
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.)
Micron Technology Inc
Original Assignee
STMicroelectronics SRL
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 STMicroelectronics SRL filed Critical STMicroelectronics SRL
Assigned to STMICROELECTRONICS S.R.L. reassignment STMICROELECTRONICS S.R.L. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: GUIDA, NICOLA, IACULO, MASSIMO, RUGGIERO, ANDREA
Publication of US20060018166A1 publication Critical patent/US20060018166A1/en
Priority to US12/033,855 priority Critical patent/US7603593B2/en
Assigned to MICRON TECHNOLOGY, INC. reassignment MICRON TECHNOLOGY, INC. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: STMICROELECTRONICS S.R.L.
Abandoned legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G11INFORMATION STORAGE
    • G11CSTATIC STORES
    • G11C29/00Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
    • G11C29/70Masking faults in memories by using spares or by reconfiguring
    • G11C29/76Masking faults in memories by using spares or by reconfiguring using address translation or modifications

Definitions

  • the present invention relates to a method for managing bad memory blocks in a nonvolatile-memory device, and to a nonvolatile-memory device implementing the management method.
  • nonvolatile-memory devices having an architecture of a NAND type have a memory array divided into memory blocks, each of which is in turn divided into sub-blocks referred to as pages.
  • Each page is formed by a given number of bytes, the majority of which, approximately 95%, define the so-called data area, while the remaining bytes define an area reserved to the storage of given information (“page spare area”).
  • Nonvolatile-memory devices having an architecture of a NAND type can be divided into two major categories according to the number of bad memory blocks that they contain: the first category comprises so-called good nonvolatile-memory devices, which do not contain bad memory blocks, while the second category includes so-called “mostly good” storage devices, which contain a number of bad memory blocks lower than a given threshold.
  • the nonvolatile-memory devices having an architecture of a NAND type shall not intrinsically contain, when they have just been produced, a number of bad memory blocks higher than a given threshold, typically in the region of 2% of the available memory blocks.
  • some manufacturers of nonvolatile-memory devices guarantee the execution of a number of write and erase cycles that is decidedly greater than one hundred thousand during the life of the storage device, typically three hundred thousand, so that it is necessary for the number of bad memory blocks that will inevitably be created during the further two hundred thousand write and erase cycles not to be higher than another threshold, typically in the region of 1.8% of the non-bad memory blocks after the first one hundred thousand cycles.
  • the management of the bad memory blocks is performed by some manufacturers of nonvolatile-memory devices by dividing the memory array into two physically distinct areas, one reserved to the users (user-block area—UBA), and one reserved to re-mapping of bad memory blocks present in the UBA (reserved-block area—RBA).
  • UBA user-block area
  • RBA reserved-block area
  • re-mapping of bad memory blocks into spare memory blocks is performed by first constructing a re-directing table, which, for each bad memory block, stores the address of the bad memory block and the address of the corresponding spare memory block, and then re-directing any request for access to said bad memory blocks to the corresponding spare memory blocks.
  • the re-directing table is then stored permanently in the area of memory reserved to the users and is read whenever the user requests access to a memory block. It is then verified whether the address of the memory block to which access is requested is contained in the re-directing table and, if it is, the address of the spare memory block in which said bad memory block has been re-mapped is read, and the access is then re-directed towards said spare memory block.
  • the re-directing table involves a rather high occupation of memory, so that its storage typically entails a considerable reduction in the amount of memory actually available to users.
  • One aim of the present invention is to provide a method for managing bad memory blocks in a nonvolatile-memory device, implementation of which will call for a reduced occupation of memory.
  • a method for managing bad memory blocks in a nonvolatile-memory device comprising a plurality of available memory blocks, each of which associated to a respective logic address
  • said management method comprising the steps of: dividing the available memory blocks into a first set and a second set, the first set being formed by addressable memory blocks that are to be used by a user, and the second set being formed by spare memory blocks reserved to the replacement of bad addressable memory blocks; each memory block assuming, in the respective set, a respective position; and re-mapping the bad addressable memory blocks in corresponding spare memory blocks; said step of re-mapping comprising the step of: constructing a data base containing, for each bad addressable memory block, information indicating the logic address of the corresponding spare memory block; said management method being characterized in that said step of creating a data base comprises the steps of: seeking bad spare memory blocks; storing the logic address of each bad spare memory block in a re-directing vector in a position corresponding to that of the
  • a nonvolatile-memory device comprising a plurality of memory blocks, and means for managing bad memory blocks
  • said management means are configured in such a way as to implement a management method comprising the steps of: dividing the available memory blocks into a first set and a second set, the first set being formed by addressable memory blocks that are to be used by a user, and the second set being formed by spare memory blocks reserved to the replacement of bad addressable memory blocks; each memory block assuming, in the respective set, a respective position; and re-mapping the bad addressable memory blocks in corresponding spare memory blocks; said step of re-mapping comprising the step of: constructing a data base containing, for each bad addressable memory block, information indicating the logic address of the corresponding spare memory block; said step of creating a data base comprises the steps of: seeking bad spare memory blocks; storing the logic address of each bad spare memory block in a re-directing vector in a position corresponding to that of the
  • FIG. 1 shows a flowchart of the management of the re-mapping of bad memory blocks according to the present invention
  • FIG. 2 is a schematic illustration of the memory blocks of a nonvolatile-memory device and a corresponding re-directing vector according to the invention
  • FIGS. 3 and 4 show how the re-directing vector of FIG. 1 is modified when a new memory block becomes bad during operation of the storage device
  • FIG. 5 shows a flowchart of the management of a request for access to a bad memory block.
  • FIG. 1 shows a flowchart of the re-mapping of bad memory blocks according to the present invention.
  • FIG. 2 is a schematic illustration of the memory blocks of a nonvolatile-memory device and the respective addresses, as well as a re-directing vector according to the present invention.
  • the non-bad memory blocks are designated by a white background, whilst the bad memory blocks are designated by a dotted background.
  • the available memory blocks are logically divided into two sets, one formed by memory blocks reserved to users, which, for simplicity, are referred hereinafter to as addressable memory blocks, and the other formed by memory blocks reserved to the replacement of bad addressable memory blocks, which are designated hereinafter, for simplicity, as “spare memory blocks” (block 100 ).
  • associated with the available memory blocks are respective logic addresses, which increase from a minimum value equal to one up to a maximum value equal to the total number of available memory blocks, which obviously depends upon the total storage capacity of the nonvolatile-memory device, and the aforementioned division is hence performed by calculating a boundary address between the two sets such that the number of spare memory blocks will satisfy the aforementioned specifications corresponding to the number of write and erase operations of the addressable memory blocks.
  • the boundary address is chosen so that the addressable memory blocks will have lower addresses than the boundary address, while the spare memory blocks will have addresses greater than or equal to the boundary address, i.e., so that the addressable memory blocks will have addresses increasing from a minimum address equal to one up to a maximum address equal to the boundary address less one, while the spare memory blocks will have addresses increasing from a minimum address equal to the boundary address up to a maximum address depending upon the overall storage capacity of the nonvolatile-memory device.
  • the memory blocks are arranged within the corresponding set according to an increasing logical order that defines their respective positions within the set; i.e., the memory blocks assume, within the corresponding set, logical positions increasing from a minimum value equal to one up to a maximum value equal to the number of memory blocks forming the set.
  • a memory block can become bad either during programming of any of the pages of the memory block or during erasing of the entire memory block, and is generally marked as such, simply by setting at a given value, which for design specifications is other than FF in hexadecimal, a given byte, typically the sixth, of the reserved area of a given page, typically the first or the second, of the memory block.
  • the verification of the “badness” of a spare memory block is thus performed by simply verifying whether the sixth byte of the spare area of the first page or second page of the memory block is different from FF in hexadecimal.
  • the spare memory blocks are examined in an increasing way, i.e., starting from the one having an address equal to the boundary address up to the one having maximum address, and whenever a bad spare memory block is identified, its address is stored in a re-directing vector in the position corresponding to the one assumed by the bad spare memory block in the respective set (block 120 ).
  • the address of the bad spare memory block is stored in the re-directing vector in the position corresponding to the offset that the address of the bad spare memory block has with respect to the boundary address, increased by one, i.e., in other words, corresponding to the difference between the address of the bad spare memory block and the boundary address, plus one.
  • the bad spare memory blocks are the second, i.e., the one with address 3942
  • the fourth i.e., the one with address 3944
  • their addresses are stored, respectively, in the second and in the fourth position of the re-directing vector, which correspond precisely to the positions assumed by the spare memory blocks in the respective set.
  • bad addressable memory blocks are then sought (block 130 ). The verification of the badness of an addressable memory block is performed similarly to what has been described above regarding to the spare memory blocks.
  • the addressable memory blocks are examined in an increasing way, starting from the one having an address equal to one up to the one having the address immediately preceding the boundary address, and whenever a bad addressable memory block is identified, its address is stored in a free position, i.e., one not occupied by the address of a bad spare memory block, of the re-directing vector (block 140 ).
  • Each bad addressable memory block is thus re-mapped into a corresponding spare memory block having a position, in the corresponding set, corresponding to that of the address of the bad addressable memory block in the re-directing vector.
  • the bad addressable memory block is re-mapped into the spare memory block having an address equal to the sum of the boundary address and of the position of the address of the bad addressable memory block in the re-directing vector, plus one.
  • re-mapping involves, in addition to re-directing, also the transfer of this content into the corresponding spare memory block (block 150 ).
  • the bad addressable memory blocks are the third, i.e., the one with address 3
  • the fifth i.e., the one with address 5
  • their addresses are stored in the first and, respectively, in the third position of the re-directing vector, which are precisely the first two free positions of the re-directing vector.
  • bad memory blocks are thus re-mapped into the first spare memory block, i.e., the one with address 3491 , and in the third spare memory block, i.e., the one with address 3493 , the positions whereof in the respective set correspond precisely to the positions of their addresses in the re-directing vector.
  • the re-directing vector thus constructed is finally stored in the storage device in one of the addressable memory blocks, for example in the last, i.e., the one having the address preceding the boundary address (block 160 ) in such a way as to enable it to be read whenever the user requests access to an addressable memory block.
  • FIG. 3 shows how the re-directing vector is modified when a new addressable memory block becomes bad during operation of the storage device, assuming as initial condition the one illustrated in FIG. 2 .
  • the address of the addressable memory block that has become bad in the example considered the fourth, i.e., the one with address 4 , is simply stored in the first free position of the re-directing vector, in the example considered the fifth position.
  • the addressable memory block that has become bad is re-mapped into the corresponding spare memory block, the position whereof in the respective set corresponds to the position of the address of the new bad addressable memory block in the re-directing vector, in the example considered the fifth spare memory block, with address 3945 .
  • FIG. 4 shows, instead, how the re-directing vector is modified when it is a spare memory block, that becomes bad during operation of the storage device, assuming as initial condition the one illustrated in FIG. 3 .
  • the address of the spare memory block that has become bad in the example the first, i.e., the one with address 3941 , is stored in the re-directing vector in the position corresponding to that of the spare memory block that has become bad in the corresponding set, in the example considered the first position. If in the re-directing vector this position is already occupied by the address of a bad addressable memory block, as in the example illustrated in FIG. 4 , then this position is freed by displacing the address stored therein into the first free position of the re-directing vector, in the example considered the sixth position.
  • the addressable memory block previously re-mapped into the bad spare memory block is thus re-mapped into the new spare memory block having, in the respective set, a position corresponding to the one into which the address of the bad addressable memory block has been shifted, in the example considered in the sixth spare memory block, i.e., the one with address 3946 .
  • Storage of the addresses of the bad memory blocks in the re-directing vector performed in the way described above thus defines a strict biunique correspondence between the positions of the spare memory blocks, in which the bad addressable memory blocks are re-mapped, and the positions in which the addresses of the bad addressable memory blocks are stored within the re-directing vector.
  • the above biunique correspondence enables re-mapping of the bad memory blocks without having to store the addresses of the spare memory blocks in which the bad addressable memory blocks are re-mapped, but by constructing a simple one-dimensional re-directing vector the memory occupation whereof is approximately half that of the re-directing table according to the known art.
  • FIG. 5 shows a flowchart corresponding to the management of a request for access to a bad addressable memory block.
  • the stored re-directing vector is first read (block 200 ), then it is verified whether the address of this addressable memory block is contained in the re-directing vector (block 210 ).
  • the function of search for the address of an addressable memory block within the re-directing vector can be performed via a simple ordered pointer, and rendered fast by the adoption of a hardware accelerator.
  • the division of the available memory blocks into spare memory blocks and addressable memory blocks could be performed in a different way from the one described.
  • the spare memory blocks could be the ones having addresses smaller than or equal to the boundary address and the addressable memory blocks could be the ones having addresses greater than the boundary address, or else the spare memory blocks could be the ones having addresses comprised between a lower boundary address and a higher boundary address.
  • the memory blocks could be arranged within the respective sets according to different orders from what has been described; for example, they could be arranged in decreasing order or else according to any other logical order that in any case defines one of its logical positions within the set.
  • the criterion with which the addresses of the bad addressable memory blocks are stored in the re-directing vector could be different from what has been described, for example the reverse, i.e., the addresses could be stored starting from the highest free positions down to the lowest.

Abstract

A method for managing bad memory blocks of a nonvolatile-memory device, in which the available memory blocks are divided into a first set, formed by addressable memory blocks that are to be used by a user, and a second set, formed by spare memory blocks that are to replace bad addressable memory blocks, and in which the bad addressable memory blocks are re-mapped into corresponding spare memory blocks. The re-mapping of the bad addressable memory blocks envisages: seeking bad spare memory blocks; storing the logic address of each bad spare memory block in a re-directing vector in a position corresponding to that of the bad spare memory block in the respective set; seeking bad addressable memory blocks; and storing the logic address of each bad addressable memory block in a free position in the re-directing vector.

Description

    BACKGROUND OF THE INVENTION
  • 1. Field of the Invention
  • The present invention relates to a method for managing bad memory blocks in a nonvolatile-memory device, and to a nonvolatile-memory device implementing the management method.
  • 2. Discussion of Related Art
  • As is known, nonvolatile-memory devices having an architecture of a NAND type have a memory array divided into memory blocks, each of which is in turn divided into sub-blocks referred to as pages. Each page is formed by a given number of bytes, the majority of which, approximately 95%, define the so-called data area, while the remaining bytes define an area reserved to the storage of given information (“page spare area”).
  • Nonvolatile-memory devices having an architecture of a NAND type can be divided into two major categories according to the number of bad memory blocks that they contain: the first category comprises so-called good nonvolatile-memory devices, which do not contain bad memory blocks, while the second category includes so-called “mostly good” storage devices, which contain a number of bad memory blocks lower than a given threshold.
  • For the purpose of guaranteeing the execution of a minimum number of write and erase cycles during the life of the storage device, typically one hundred thousand, the nonvolatile-memory devices having an architecture of a NAND type shall not intrinsically contain, when they have just been produced, a number of bad memory blocks higher than a given threshold, typically in the region of 2% of the available memory blocks.
  • In response to the continuous market requirements, some manufacturers of nonvolatile-memory devices guarantee the execution of a number of write and erase cycles that is decidedly greater than one hundred thousand during the life of the storage device, typically three hundred thousand, so that it is necessary for the number of bad memory blocks that will inevitably be created during the further two hundred thousand write and erase cycles not to be higher than another threshold, typically in the region of 1.8% of the non-bad memory blocks after the first one hundred thousand cycles.
  • To manage the presence of bad memory blocks, it is known practice to provide, in nonvolatile-memory devices, a certain number of spare memory blocks, in which the bad memory blocks are re-mapped.
  • For example, in a 512-Mbit flash memory device, with 4096 memory blocks, each formed by 32 pages of 528 bytes each, of which 512 defining the data area and the remaining 16 defining the reserved area, it is necessary to provide a number of spare memory blocks equal to 153, of which 82 have the function of guaranteeing reaching the first one hundred thousand write and erase cycles, and the other 71 have the function of guaranteeing reaching three hundred thousand write and erase cycles.
  • The management of the bad memory blocks is performed by some manufacturers of nonvolatile-memory devices by dividing the memory array into two physically distinct areas, one reserved to the users (user-block area—UBA), and one reserved to re-mapping of bad memory blocks present in the UBA (reserved-block area—RBA).
  • In particular, re-mapping of bad memory blocks into spare memory blocks is performed by first constructing a re-directing table, which, for each bad memory block, stores the address of the bad memory block and the address of the corresponding spare memory block, and then re-directing any request for access to said bad memory blocks to the corresponding spare memory blocks.
  • The re-directing table is then stored permanently in the area of memory reserved to the users and is read whenever the user requests access to a memory block. It is then verified whether the address of the memory block to which access is requested is contained in the re-directing table and, if it is, the address of the spare memory block in which said bad memory block has been re-mapped is read, and the access is then re-directed towards said spare memory block.
  • Generally, however, the re-directing table involves a rather high occupation of memory, so that its storage typically entails a considerable reduction in the amount of memory actually available to users.
  • SUMMARY OF THE INVENTION
  • One aim of the present invention is to provide a method for managing bad memory blocks in a nonvolatile-memory device, implementation of which will call for a reduced occupation of memory.
  • According to the present invention a method is provided for managing bad memory blocks in a nonvolatile-memory device, comprising a plurality of available memory blocks, each of which associated to a respective logic address, said management method comprising the steps of: dividing the available memory blocks into a first set and a second set, the first set being formed by addressable memory blocks that are to be used by a user, and the second set being formed by spare memory blocks reserved to the replacement of bad addressable memory blocks; each memory block assuming, in the respective set, a respective position; and re-mapping the bad addressable memory blocks in corresponding spare memory blocks; said step of re-mapping comprising the step of: constructing a data base containing, for each bad addressable memory block, information indicating the logic address of the corresponding spare memory block; said management method being characterized in that said step of creating a data base comprises the steps of: seeking bad spare memory blocks; storing the logic address of each bad spare memory block in a re-directing vector in a position corresponding to that of the bad spare memory block in the respective set; seeking bad addressable memory blocks; and storing the logic address of each bad addressable memory block in a free position in the re-directing vector.
  • Moreover, according to the present invention, there is provided a nonvolatile-memory device, comprising a plurality of memory blocks, and means for managing bad memory blocks, wherein said management means are configured in such a way as to implement a management method comprising the steps of: dividing the available memory blocks into a first set and a second set, the first set being formed by addressable memory blocks that are to be used by a user, and the second set being formed by spare memory blocks reserved to the replacement of bad addressable memory blocks; each memory block assuming, in the respective set, a respective position; and re-mapping the bad addressable memory blocks in corresponding spare memory blocks; said step of re-mapping comprising the step of: constructing a data base containing, for each bad addressable memory block, information indicating the logic address of the corresponding spare memory block; said step of creating a data base comprises the steps of: seeking bad spare memory blocks; storing the logic address of each bad spare memory block in a re-directing vector in a position corresponding to that of the bad spare memory block in the respective set; seeking bad addressable memory blocks; and storing the logic address of each bad addressable memory block in a free position in the re-directing vector.
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • For a better understanding of the present invention, a preferred embodiment thereof is described hereinafter, purely as a non-limiting example, with reference to the attached drawings, wherein:
  • FIG. 1 shows a flowchart of the management of the re-mapping of bad memory blocks according to the present invention;
  • FIG. 2 is a schematic illustration of the memory blocks of a nonvolatile-memory device and a corresponding re-directing vector according to the invention;
  • FIGS. 3 and 4 show how the re-directing vector of FIG. 1 is modified when a new memory block becomes bad during operation of the storage device; and
  • FIG. 5 shows a flowchart of the management of a request for access to a bad memory block.
  • DETAILED DESCRIPTION
  • FIG. 1 shows a flowchart of the re-mapping of bad memory blocks according to the present invention. To facilitate understanding of the present invention, the following description refers also to the example illustrated in FIG. 2, which is a schematic illustration of the memory blocks of a nonvolatile-memory device and the respective addresses, as well as a re-directing vector according to the present invention. In FIG. 2, moreover, the non-bad memory blocks are designated by a white background, whilst the bad memory blocks are designated by a dotted background.
  • As illustrated in FIG. 1, initially the available memory blocks are logically divided into two sets, one formed by memory blocks reserved to users, which, for simplicity, are referred hereinafter to as addressable memory blocks, and the other formed by memory blocks reserved to the replacement of bad addressable memory blocks, which are designated hereinafter, for simplicity, as “spare memory blocks” (block 100).
  • In particular, associated with the available memory blocks are respective logic addresses, which increase from a minimum value equal to one up to a maximum value equal to the total number of available memory blocks, which obviously depends upon the total storage capacity of the nonvolatile-memory device, and the aforementioned division is hence performed by calculating a boundary address between the two sets such that the number of spare memory blocks will satisfy the aforementioned specifications corresponding to the number of write and erase operations of the addressable memory blocks.
  • According to one embodiment, the boundary address is chosen so that the addressable memory blocks will have lower addresses than the boundary address, while the spare memory blocks will have addresses greater than or equal to the boundary address, i.e., so that the addressable memory blocks will have addresses increasing from a minimum address equal to one up to a maximum address equal to the boundary address less one, while the spare memory blocks will have addresses increasing from a minimum address equal to the boundary address up to a maximum address depending upon the overall storage capacity of the nonvolatile-memory device.
  • In this way, the memory blocks are arranged within the corresponding set according to an increasing logical order that defines their respective positions within the set; i.e., the memory blocks assume, within the corresponding set, logical positions increasing from a minimum value equal to one up to a maximum value equal to the number of memory blocks forming the set.
  • With reference once again to FIG. 1, after determining the boundary address, possible bad spare memory blocks are sought (block 110).
  • Generally, a memory block can become bad either during programming of any of the pages of the memory block or during erasing of the entire memory block, and is generally marked as such, simply by setting at a given value, which for design specifications is other than FF in hexadecimal, a given byte, typically the sixth, of the reserved area of a given page, typically the first or the second, of the memory block. The verification of the “badness” of a spare memory block is thus performed by simply verifying whether the sixth byte of the spare area of the first page or second page of the memory block is different from FF in hexadecimal.
  • In particular, the spare memory blocks are examined in an increasing way, i.e., starting from the one having an address equal to the boundary address up to the one having maximum address, and whenever a bad spare memory block is identified, its address is stored in a re-directing vector in the position corresponding to the one assumed by the bad spare memory block in the respective set (block 120). In mathematical terms, the address of the bad spare memory block is stored in the re-directing vector in the position corresponding to the offset that the address of the bad spare memory block has with respect to the boundary address, increased by one, i.e., in other words, corresponding to the difference between the address of the bad spare memory block and the boundary address, plus one.
  • The above is illustrated in the example of FIG. 2, where the bad spare memory blocks are the second, i.e., the one with address 3942, and the fourth, i.e., the one with address 3944, and their addresses are stored, respectively, in the second and in the fourth position of the re-directing vector, which correspond precisely to the positions assumed by the spare memory blocks in the respective set.
  • With reference once again to FIG. 1, once the search for the bad spare memory blocks is terminated, bad addressable memory blocks are then sought (block 130). The verification of the badness of an addressable memory block is performed similarly to what has been described above regarding to the spare memory blocks.
  • In particular, the addressable memory blocks are examined in an increasing way, starting from the one having an address equal to one up to the one having the address immediately preceding the boundary address, and whenever a bad addressable memory block is identified, its address is stored in a free position, i.e., one not occupied by the address of a bad spare memory block, of the re-directing vector (block 140).
  • Conveniently, storage of the addresses of the addressable memory blocks identified as being bad is performed starting from the lowest free positions up to the highest ones.
  • Each bad addressable memory block is thus re-mapped into a corresponding spare memory block having a position, in the corresponding set, corresponding to that of the address of the bad addressable memory block in the re-directing vector. In more precise terms, the bad addressable memory block is re-mapped into the spare memory block having an address equal to the sum of the boundary address and of the position of the address of the bad addressable memory block in the re-directing vector, plus one.
  • Should a bad memory block have an own content, re-mapping involves, in addition to re-directing, also the transfer of this content into the corresponding spare memory block (block 150).
  • The above is illustrated in the example of FIG. 2, where the bad addressable memory blocks are the third, i.e., the one with address 3, and the fifth, i.e., the one with address 5, and their addresses are stored in the first and, respectively, in the third position of the re-directing vector, which are precisely the first two free positions of the re-directing vector.
  • These bad memory blocks are thus re-mapped into the first spare memory block, i.e., the one with address 3491, and in the third spare memory block, i.e., the one with address 3493, the positions whereof in the respective set correspond precisely to the positions of their addresses in the re-directing vector.
  • With reference once again to FIG. 1, the re-directing vector thus constructed is finally stored in the storage device in one of the addressable memory blocks, for example in the last, i.e., the one having the address preceding the boundary address (block 160) in such a way as to enable it to be read whenever the user requests access to an addressable memory block.
  • FIG. 3 shows how the re-directing vector is modified when a new addressable memory block becomes bad during operation of the storage device, assuming as initial condition the one illustrated in FIG. 2.
  • In a way similar to the above, the address of the addressable memory block that has become bad, in the example considered the fourth, i.e., the one with address 4, is simply stored in the first free position of the re-directing vector, in the example considered the fifth position. In this way, then, the addressable memory block that has become bad is re-mapped into the corresponding spare memory block, the position whereof in the respective set corresponds to the position of the address of the new bad addressable memory block in the re-directing vector, in the example considered the fifth spare memory block, with address 3945.
  • FIG. 4 shows, instead, how the re-directing vector is modified when it is a spare memory block, that becomes bad during operation of the storage device, assuming as initial condition the one illustrated in FIG. 3.
  • In a way similar to what the above, the address of the spare memory block that has become bad, in the example the first, i.e., the one with address 3941, is stored in the re-directing vector in the position corresponding to that of the spare memory block that has become bad in the corresponding set, in the example considered the first position. If in the re-directing vector this position is already occupied by the address of a bad addressable memory block, as in the example illustrated in FIG. 4, then this position is freed by displacing the address stored therein into the first free position of the re-directing vector, in the example considered the sixth position. The addressable memory block previously re-mapped into the bad spare memory block is thus re-mapped into the new spare memory block having, in the respective set, a position corresponding to the one into which the address of the bad addressable memory block has been shifted, in the example considered in the sixth spare memory block, i.e., the one with address 3946.
  • Storage of the addresses of the bad memory blocks in the re-directing vector performed in the way described above thus defines a strict biunique correspondence between the positions of the spare memory blocks, in which the bad addressable memory blocks are re-mapped, and the positions in which the addresses of the bad addressable memory blocks are stored within the re-directing vector.
  • The above biunique correspondence enables re-mapping of the bad memory blocks without having to store the addresses of the spare memory blocks in which the bad addressable memory blocks are re-mapped, but by constructing a simple one-dimensional re-directing vector the memory occupation whereof is approximately half that of the re-directing table according to the known art.
  • Finally, FIG. 5 shows a flowchart corresponding to the management of a request for access to a bad addressable memory block.
  • According to FIG. 5, when a user requests access to an addressable memory block, for example to read or write data there, the stored re-directing vector is first read (block 200), then it is verified whether the address of this addressable memory block is contained in the re-directing vector (block 210).
  • If it is not (output NO from block 100), then access to this addressable memory block is granted (block 220); otherwise (output YES from block 100), the position within the re-directing vector in which its address has been stored is identified (block 230), the address of the spare memory block in which the bad addressable memory block has been re-mapped is determined (block 240), and finally access is re-directed to the spare memory block (block 250).
  • In particular, thanks to the biunique correspondence existing between the position, within the corresponding set, of the spare memory block in which the bad addressable memory block has been re-mapped and the position, within the re-directing vector, of the address of the bad addressable memory block, to determine the address of the spare memory block it is sufficient simply to sum up the boundary address and the position of the address of the bad addressable memory block in the re-directing vector, and decrease the result by one.
  • From an examination of the characteristics of the management method according to the present invention the advantages that it makes possible are evident.
  • In particular, via a simple re-directing vector constructed in the way described above it is possible to store which are the bad memory blocks, where these memory blocks are re-mapped and where a new bad memory block can be re-mapped. In particular, thanks to the biunique correspondence existing between the positions of the spare memory blocks in the respective set and the positions of the addresses of the bad addressable memory blocks in the re-directing vector, the re-mapping of the bad memory blocks is performed, without having to store the addresses of the spare memory blocks in which the bad memory blocks have been re-mapped, with a saving in terms of occupation of memory of approximately 50% with respect to the known art.
  • Furthermore, the function of search for the address of an addressable memory block within the re-directing vector can be performed via a simple ordered pointer, and rendered fast by the adoption of a hardware accelerator.
  • Finally, it is clear that modifications and variations can be made to the method described and illustrated herein, without thereby departing from the scope of protection of the present invention, as defined in the annexed claims.
  • In particular, the division of the available memory blocks into spare memory blocks and addressable memory blocks could be performed in a different way from the one described. For example, the spare memory blocks could be the ones having addresses smaller than or equal to the boundary address and the addressable memory blocks could be the ones having addresses greater than the boundary address, or else the spare memory blocks could be the ones having addresses comprised between a lower boundary address and a higher boundary address.
  • Furthermore, the memory blocks could be arranged within the respective sets according to different orders from what has been described; for example, they could be arranged in decreasing order or else according to any other logical order that in any case defines one of its logical positions within the set.
  • Finally, the criterion with which the addresses of the bad addressable memory blocks are stored in the re-directing vector could be different from what has been described, for example the reverse, i.e., the addresses could be stored starting from the highest free positions down to the lowest.
  • Having thus described at least one illustrative embodiment of the invention, various alterations, modifications, and improvements will readily occur to those skilled in the art. Such alterations, modifications, and improvements are intended to be within the spirit and scope of the invention. Accordingly, the foregoing description is by way of example only and is not intended as limiting. The invention is limited only as defined in the following claims and the equivalents thereto.

Claims (6)

1. A method for managing bad memory blocks in a nonvolatile-memory device, said nonvolatile-memory device comprising a plurality of available memory blocks, each of which associated to a respective logic address, said management method comprising the steps of:
dividing the available memory blocks into a first set and a second set, the first set being formed by addressable memory blocks that are to be used by a user, and the second set being formed by spare memory blocks reserved to the replacement of bad addressable memory blocks; each memory block assuming, in the respective set, a respective position; and
re-mapping the bad addressable memory blocks in corresponding spare memory blocks;
said step of re-mapping comprising the step of:
constructing a data base containing, for each bad addressable memory block, information indicating the logic address of the corresponding spare memory block;
said step of creating a data base comprises the steps of:
seeking bad spare memory blocks;
storing the logic address of each bad spare memory block in a re-directing vector in a position corresponding to that of the bad spare memory block in the respective set;
seeking bad addressable memory blocks; and
storing the logic address of each bad addressable memory block in a free position in the re-directing vector.
2. The management method according to claim 1, further comprising the steps of:
detecting a request for access to an addressable memory block;
verifying whether the address of said addressable memory block is contained in the re-directing vector;
if the address of the addressable memory block is not contained in the re-directing vector, enabling access to said addressable memory block;
if the address of the addressable memory block is contained in the re-directing vector, reading the position of its address in the re-directing vector, and re-directing access to the spare memory block, which has, in the respective set, a position corresponding to that of the address of the addressable memory block in the re-directing vector.
3. The management method according to claim 1, wherein the step of storing the logic address of a bad spare memory block in the re-directing vector in a position corresponding to that of the bad spare memory block in the respective set comprises the steps of:
verifying whether in the re-directing vector said position is already occupied by the address of a bad addressable memory block;
if said position is free, storing the logic address of the bad spare memory block in said free position;
if said position is occupied by the address of a bad addressable memory block, shifting said address into a free position in the re-directing vector, and storing the logic address of the bad spare memory block in the position thus freed.
4. The management method according claim 1, for a nonvolatile-memory device wherein each memory block is divided into a plurality of pages, each page being formed by a plurality of bytes, wherein the step of seeking bad memory blocks comprises the step of:
verifying whether a specific byte of a specific page of a bad memory block assumes a value different from a given reference value.
5. The management method according to claim 1, in which the logic addresses of the bad addressable memory blocks are stored in the re-directing vector starting from the lowest positions.
6. A nonvolatile-memory device comprising a plurality of memory blocks, and means for managing bad memory blocks, wherein said management means are configured in such a way as to implement a management method comprising the steps of:
dividing the available memory blocks into a first set and a second set, the first set being formed by addressable memory blocks that are to be used by a user, and the second set being formed by spare memory blocks reserved to the replacement of bad addressable memory blocks; each memory block assuming, in the respective set, a respective position; and
re-mapping the bad addressable memory blocks in corresponding spare memory blocks;
said step of re-mapping comprising the step of:
constructing a data base containing, for each bad addressable memory block, information indicating the logic address of the corresponding spare memory block;
said step of creating a data base comprises the steps of:
seeking bad spare memory blocks;
storing the logic address of each bad spare memory block in a re-directing vector in a position corresponding to that of the bad spare memory block in the respective set;
seeking bad addressable memory blocks; and
storing the logic address of each bad addressable memory block in a free position in the re-directing vector.
US11/152,675 2004-06-14 2005-06-14 Method and managing bad memory blocks in a nonvolatile memory device, and nonvolatile-memory device implementing the management method Abandoned US20060018166A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US12/033,855 US7603593B2 (en) 2004-06-14 2008-02-19 Method for managing bad memory blocks in a nonvolatile-memory device, and nonvolatile-memory device implementing the management method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
EP04425435.7 2004-06-14
EP04425435A EP1607984B1 (en) 2004-06-14 2004-06-14 Method for managing bad memory blocks in a nonvolatile memory device, and nonvolatile memory device implementing the management method

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US12/033,855 Continuation US7603593B2 (en) 2004-06-14 2008-02-19 Method for managing bad memory blocks in a nonvolatile-memory device, and nonvolatile-memory device implementing the management method

Publications (1)

Publication Number Publication Date
US20060018166A1 true US20060018166A1 (en) 2006-01-26

Family

ID=34932559

Family Applications (2)

Application Number Title Priority Date Filing Date
US11/152,675 Abandoned US20060018166A1 (en) 2004-06-14 2005-06-14 Method and managing bad memory blocks in a nonvolatile memory device, and nonvolatile-memory device implementing the management method
US12/033,855 Active 2025-09-04 US7603593B2 (en) 2004-06-14 2008-02-19 Method for managing bad memory blocks in a nonvolatile-memory device, and nonvolatile-memory device implementing the management method

Family Applications After (1)

Application Number Title Priority Date Filing Date
US12/033,855 Active 2025-09-04 US7603593B2 (en) 2004-06-14 2008-02-19 Method for managing bad memory blocks in a nonvolatile-memory device, and nonvolatile-memory device implementing the management method

Country Status (3)

Country Link
US (2) US20060018166A1 (en)
EP (1) EP1607984B1 (en)
DE (1) DE602004008240T2 (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080025095A1 (en) * 2006-07-26 2008-01-31 Se-Jin Ahn Flash memory device and program method thereof
US20080270675A1 (en) * 2007-04-25 2008-10-30 Dheemanth Nagaraj Defect management for a semiconductor memory system
US7694093B2 (en) 2007-04-27 2010-04-06 Hewlett-Packard Development Company, L.P. Memory module and method for mirroring data by rank
US20100085820A1 (en) * 2008-01-28 2010-04-08 Kabushiki Kaisha Toshiba Semiconductor memory device
US20100103738A1 (en) * 2008-10-29 2010-04-29 Macronix International Co., Ltd. Memory and operating method thereof
US20110219260A1 (en) * 2007-01-22 2011-09-08 Micron Technology, Inc. Defective memory block remapping method and system, and memory device and processor-based system using same
US10146604B2 (en) * 2016-08-23 2018-12-04 Oracle International Corporation Bad block detection and predictive analytics in NAND flash storage devices
US10339020B2 (en) * 2016-03-29 2019-07-02 Toshiba Memory Corporation Object storage system, controller and storage medium
US10445199B2 (en) * 2016-12-22 2019-10-15 Western Digital Technologies, Inc. Bad page management in storage devices

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI298836B (en) * 2005-10-12 2008-07-11 Sunplus Technology Co Ltd Apparatus for controlling flash memory and method thereof
US7925939B2 (en) * 2008-09-26 2011-04-12 Macronix International Co., Ltd Pre-code device, and pre-code system and pre-coding method thererof
US8839053B2 (en) * 2010-05-27 2014-09-16 Microsoft Corporation Error correcting pointers for non-volatile storage
US9032244B2 (en) 2012-11-16 2015-05-12 Microsoft Technology Licensing, Llc Memory segment remapping to address fragmentation
KR102136396B1 (en) 2013-08-30 2020-07-22 삼성전자주식회사 DRAM controller having bad pages management function and therefore bad pages management method
US9569322B2 (en) * 2014-03-31 2017-02-14 Oracle International Corporation Memory migration in presence of live memory traffic
US10223018B2 (en) 2017-04-19 2019-03-05 Sandisk Technologies Llc Bad page and bad block management in memory

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4827452A (en) * 1985-08-13 1989-05-02 Mitsubishi Denki Kabushiki Kaisha Semiconductor memory including a selectively disabled redunancy circuit
US5781717A (en) * 1996-09-19 1998-07-14 I-Cube, Inc. Dynamic spare column replacement memory system
US6567950B1 (en) * 1999-04-30 2003-05-20 International Business Machines Corporation Dynamically replacing a failed chip

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2136992A (en) * 1983-03-18 1984-09-26 Georg V Coza Method and System of Ensuring Integrity of Data in an Electronic Memory
JPH03137744A (en) 1989-10-24 1991-06-12 Toshiba Corp Memory control system
WO1992020068A1 (en) 1991-05-07 1992-11-12 Sophos Technologic Fast memory system employing mostly good memories

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4827452A (en) * 1985-08-13 1989-05-02 Mitsubishi Denki Kabushiki Kaisha Semiconductor memory including a selectively disabled redunancy circuit
US5781717A (en) * 1996-09-19 1998-07-14 I-Cube, Inc. Dynamic spare column replacement memory system
US6567950B1 (en) * 1999-04-30 2003-05-20 International Business Machines Corporation Dynamically replacing a failed chip

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080025095A1 (en) * 2006-07-26 2008-01-31 Se-Jin Ahn Flash memory device and program method thereof
US20110219260A1 (en) * 2007-01-22 2011-09-08 Micron Technology, Inc. Defective memory block remapping method and system, and memory device and processor-based system using same
US8601331B2 (en) * 2007-01-22 2013-12-03 Micron Technology, Inc. Defective memory block remapping method and system, and memory device and processor-based system using same
US20080270675A1 (en) * 2007-04-25 2008-10-30 Dheemanth Nagaraj Defect management for a semiconductor memory system
US7996710B2 (en) * 2007-04-25 2011-08-09 Hewlett-Packard Development Company, L.P. Defect management for a semiconductor memory system
US7694093B2 (en) 2007-04-27 2010-04-06 Hewlett-Packard Development Company, L.P. Memory module and method for mirroring data by rank
US20100085820A1 (en) * 2008-01-28 2010-04-08 Kabushiki Kaisha Toshiba Semiconductor memory device
US8644072B2 (en) * 2008-01-28 2014-02-04 Kabushiki Kaisha Toshiba Three dimensionally stacked memory and the isolation of memory cell layer
US8275929B2 (en) * 2008-10-29 2012-09-25 Macronix International Co., Ltd. Memory and operating method thereof
US20100103738A1 (en) * 2008-10-29 2010-04-29 Macronix International Co., Ltd. Memory and operating method thereof
TWI419168B (en) * 2008-10-29 2013-12-11 Macronix Int Co Ltd Data programming method and data reading method and one-time programmable memory using thereof
US10339020B2 (en) * 2016-03-29 2019-07-02 Toshiba Memory Corporation Object storage system, controller and storage medium
US10146604B2 (en) * 2016-08-23 2018-12-04 Oracle International Corporation Bad block detection and predictive analytics in NAND flash storage devices
US10445199B2 (en) * 2016-12-22 2019-10-15 Western Digital Technologies, Inc. Bad page management in storage devices

Also Published As

Publication number Publication date
US20080158962A1 (en) 2008-07-03
DE602004008240T2 (en) 2008-05-15
EP1607984A1 (en) 2005-12-21
US7603593B2 (en) 2009-10-13
DE602004008240D1 (en) 2007-09-27
EP1607984B1 (en) 2007-08-15

Similar Documents

Publication Publication Date Title
US7603593B2 (en) Method for managing bad memory blocks in a nonvolatile-memory device, and nonvolatile-memory device implementing the management method
KR100833188B1 (en) Non-volatile memory system storing data at single-level cell or multi-level cell based on the feature of data
US8180955B2 (en) Computing systems and methods for managing flash memory device
US8078797B2 (en) Increasing the memory performance of flash memory devices by writing sectors simultaneously to multiple flash memory devices
US7434122B2 (en) Flash memory device for performing bad block management and method of performing bad block management of flash memory device
US6202138B1 (en) Increasing the memory performance of flash memory devices by writing sectors simultaneously to multiple flash memory devices
US8332576B2 (en) Data reading method for flash memory and controller and storage system using the same
US7702844B2 (en) Address mapping method and mapping information managing method for flash memory, and flash memory using the same
US8176236B2 (en) Memory controller, memory system with memory controller, and method of controlling flash memory
US8356136B2 (en) Block management method of a non-volatile memory
US8433844B2 (en) Method for managing a memory device having multiple channels and multiple ways, and associated memory device and controller thereof
US8706998B2 (en) Method for managing flash memories having mixed memory types
US20080098195A1 (en) Memory system including flash memory and mapping table management method
US11288019B2 (en) Memory management method and storage controller
US20080120488A1 (en) Apparatus and method of managing nonvolatile memory
US8037236B2 (en) Flash memory writing method and storage system and controller using the same
JP5093294B2 (en) MEMORY CONTROLLER, FLASH MEMORY SYSTEM HAVING MEMORY CONTROLLER, AND FLASH MEMORY CONTROL METHOD
US9009442B2 (en) Data writing method, memory controller and memory storage apparatus
US20140089566A1 (en) Data storing method, and memory controller and memory storage apparatus using the same
US9778862B2 (en) Data storing method for preventing data losing during flush operation, memory control circuit unit and memory storage apparatus
JP4636005B2 (en) MEMORY CONTROLLER, FLASH MEMORY SYSTEM HAVING MEMORY CONTROLLER, AND FLASH MEMORY CONTROL METHOD
US9830077B2 (en) Data writing method, memory control circuit unit and memory storage apparatus
KR100780963B1 (en) Memory card and driving method thereof
JP4241741B2 (en) Memory controller and flash memory system
CN110289036B (en) Read voltage optimization method and memory controller

Legal Events

Date Code Title Description
AS Assignment

Owner name: STMICROELECTRONICS S.R.L., ITALY

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:IACULO, MASSIMO;GUIDA, NICOLA;RUGGIERO, ANDREA;REEL/FRAME:017092/0721

Effective date: 20050912

STCB Information on status: application discontinuation

Free format text: ABANDONED -- FAILURE TO PAY ISSUE FEE

AS Assignment

Owner name: MICRON TECHNOLOGY, INC., IDAHO

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:STMICROELECTRONICS S.R.L.;REEL/FRAME:029186/0761

Effective date: 20080206