The present disclosure relates generally to semiconductor memories and, in particular, in one or more embodiments, the present disclosure relates to a configuration of memory cells in non-volatile memory devices.
Memory devices are typically provided as internal, semiconductor, integrated circuits in computers or other electronic devices. There are many different types of memory including random-access memory (RAM), read only memory (ROM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), and flash memory.
Flash memory devices have developed into a popular source of non-volatile memory for a wide range of electronic applications. Non-volatile memory is memory that can retain its stored data for some extended period without the application of power. Common uses for flash memory and other non-volatile memory include personal computers, personal digital assistants (PDAs), digital cameras, digital media players, digital recorders, games, appliances, vehicles, wireless devices, mobile telephones and removable memory modules, and the uses for non-volatile memory continue to expand.
Flash memory is typically arranged in a memory array 100 comprising a plurality of blocks of memory cells such as shown in
Recognizing that errors might occur in the reading of data values from the memory device, the memory device might employ one or more types of error correction methods. Error Correction Schemes, commonly referred to as error correction codes (ECC), can be implemented in order to detect and/or attempt to correct these errors. Various ECCs comprise codes in which each data signal subjected to the ECC should conform to the specific rules of construction of the ECC. Departures from this construction of data (i.e., errors) that are not too great can generally be automatically detected and sometimes corrected. Examples of ECCs include Hamming code, BCH code, Reed-Solomon code, Reed-Muller code, Binary Golay code, Low-Density Parity code, and Trellis Code modulation. Some ECCs can correct single-bit errors and detect double-bit errors. Other ECCs can detect and/or correct multi-bit errors, for example.
Typically, a memory device will store user data in a first group of memory cells (which can be viewed as, for example, a first set of memory locations, a user data space, a user data portion and/or a user data storage area) and error correction code (ECC) data in a second group of memory cells of each block. These memory cell groupings are typically determined during design and layout of the memory device and the memory allocation for user data storage and ECC data storage is typically a permanent configuration of the memory device. During a read operation, both the stored user data and the ECC data are read from the memory array in response to a read request of the user data. Using known algorithms, the user data returned from the read operation is compared to the ECC data. If errors are detected and those errors are within the limits of the ECC, e.g., sufficient ECC resolution exists in the stored ECC data, the errors may be corrected.
By way of example, each block 102 of
However, different ECC codes require different amounts of memory cells to store their respective ECC data. Thus, if it is desired to accommodate more than one ECC code, the ECC data storage area of each block of the memory device is configured (e.g., allocated) to support an ECC code having the largest ECC storage requirements that might be utilized. However, this can be inefficient in situations wherein an ECC code is utilized which does not require the entire amount of allocated ECC data storage areas in the memory device, thus leaving memory cells unused and unavailable. A typical memory device might be allocated with blocks having 2048B (byte) of user data space and 140B of ECC data space to accommodate the largest ECC code(s) that might be desired. For example, ECC data storage areas 1121 and 1122 together might comprise 140B. This configuration (e.g., arrangement) of each block of memory would be repeated throughout the memory array 100. Thus, the memory array would comprise a number of blocks of memory each having the same user data space and ECC data space configuration per block. If an ECC is utilized which does not require the full 140B of ECC data space allocated for each block of user data memory cells, the remaining allocated ECC data space ends up not being utilized. Alternatively, if greater reliability is desired, the 140B of ECC data space might not be sufficient to store ECC data at the desired ECC resolution. For example, it may be desirable to have a particular level of reliability and a particular ECC code might be selected to be able to correct a certain number of errors occurring in user data stored in a memory device. If insufficient ECC data space is allocated, there might not be enough ECC data (e.g., insufficient ECC data resolution) to achieve the desired level of reliability and the number of errors that could be corrected might be reduced, for example.
As discussed above, the memory array configuration, such as shown in
For the reasons stated above, and for other reasons stated below which will become apparent to those skilled in the art upon reading and understanding the present disclosure, there is a need in the art for alternate configurations of memory cells in memory devices.
In the following detailed description of the invention, reference is made to the accompanying drawings that form a part hereof, and in which is shown, by way of illustration, specific embodiments in which the invention may be practiced. In the drawings, like numerals describe substantially similar components throughout the several views. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. Other embodiments may be utilized and structural, logical, and electrical changes may be made without departing from the scope of the present disclosure. The following detailed description is, therefore, not to be taken in a limiting sense.
Flash memory typically utilizes one of two basic architectures known as NOR Flash and NAND Flash. The designation is derived from the logic used to read the devices.
Memory cells are typically programmed using erase and programming cycles. For example, memory cells of a particular block of memory cells are first erased and then selectively programmed. For a NAND array, a block of memory cells is typically erased by grounding all of the word lines in the block and applying an erase voltage to a semiconductor substrate on which the block of memory cells are formed, and thus to the channels of the memory cells, in order to remove charges which might be stored on the charge storage structures (e.g., floating gates or charge traps) of the block of memory cells.
Programming typically involves applying one or more programming pulses to a selected word line (e.g., WL42184) and thus to the control gate of each memory cell 2201-4 coupled to the selected word line. Typical programming pulses start at or near 15V and tend to increase in magnitude during each programming pulse application. While the program voltage (e.g., programming pulse) is applied to the selected word line, a potential, such as a ground potential, is applied to the substrate, and thus to the channels of these memory cells, resulting in a charge transfer from the channel to the charge storage structures of memory cells targeted for programming. More specifically, the charge storage structures are typically charged through direct injection or Fowler-Nordheim tunneling of electrons from the channel to the storage structure, resulting in a Vt typically greater than zero, for example. In addition, an inhibit voltage is typically applied to bit lines not coupled to a NAND string containing a memory cell that is targeted (e.g., selected) for programming. Typically a verify operation is performed following each applied programming pulse to determine if the selected memory cells have achieved their target (e.g., intended) programmed state. A verify operation generally includes performing a sense operation to determine if a threshold voltage of a memory cell has reached a particular target value.
Memory cells (e.g., flash memory cells) can be configured as what are known in the art as Single Level Memory Cells (SLC) or Multilevel Memory Cells (MLC). SLC and MLC memory cells assign a data state (e.g., as represented by one or more bits) to a specific range of threshold voltages (Vt) stored on the memory cells. Single level memory cells (SLC) permit the storage of a single binary digit (e.g., bit) of data on each memory cell. Meanwhile, MLC technology permits the storage of two or more binary digits per cell, depending on the quantity of Vt ranges assigned to the cell and the stability of the assigned Vt ranges during the lifetime operation of the memory cell. The number of Vt ranges (e.g., levels), used to represent a bit pattern comprised of N-bits is 2N, where N is an integer. For example, one bit may be represented by two ranges, two bits by four ranges, three bits by eight ranges, etc. MLC memory cells may store even or odd numbers of bits on each memory cell, and schemes providing for fractional bits (e.g., 1.5 bits per cell) are also known. A common naming convention is to refer to SLC memory as MLC (two level) memory as SLC memory utilizes two Vt ranges in order to store one bit of data as represented by a 0 or a 1, for example. MLC memory configured to store two bits of data can be represented by MLC (four level), three bits of data by MLC (eight level), etc.
Typically there is a greater likelihood of errors occurring in data stored in higher level memory cells than in lower level memory cells. For example, MLC (four level) typically has a higher likelihood of errors than SLC memory, for example. This can be caused by the increased number and narrowing of the Vt ranges programmed in higher level memory cells. As described above, various error correcting schemes are utilized to attempt to restore data that may have been corrupted.
In contrast with the typical method of allocating user data and overhead data storage areas such as shown in
Memory device 500 further comprises row decode circuitry 510 and column decode circuitry 512 which facilitate access to the memory cells of the memory array 502. Address circuitry 508 provides addressing information to the row decode circuitry 510 and to the column decode circuitry 512. Sense circuitry and data cache 514 help facilitate memory device operations such as reading, writing (e.g., programming) and erase operations. For example, the sense circuitry might detect the programmed state of a particular number of selected memory cells to be read. Data cache 514 might store data sensed by the sense circuitry (e.g., such as sense data obtained during a read or verify operation.) Data cache 514 might also store data to be programmed into a particular number of selected memory cells, such as during a programming operation, for example.
Memory device 500 further comprises a controller (e.g., control circuitry) 506 which at least in part facilitates memory device operations according to various embodiments of the present disclosure. The control circuitry 506 might also be coupled to other circuitry of the memory device 500 (not shown), such as to the row decode circuitry 510, column decode circuitry 512 and sense circuitry/data cache 514 circuitry, for example.
In contrast with the fixed configuration of the memory array 100 as shown in
Configurations of user data storage areas and overhead data storage areas might also be different in different areas (e.g., different blocks or pages) of the memory array 502. For example, a first configuration of memory storage areas might be utilized in a first block of memory cells and different configurations might be utilized in different blocks of the memory array 502 according to various embodiments of the present disclosure. Memory storage area allocation of a particular type (e.g., user data, ECC data) might occur across one or more blocks and/or pages of memory cells. The memory storage allocation might span multiple die (i.e., multiple physical die) of the memory device according to various embodiments of the present disclosure. Spare areas might also be allocated exclusively in one or more blocks, such as at the end of the memory array 502, for example.
Control circuitry 506 might be configured to manage the configuration of the memory array 502 according to various embodiments of the present disclosure. Control circuitry 506 might be coupled to a host (e.g., processor) through an interface (not shown) as part of a memory system, for example. According to one or more embodiments, the control circuitry 506 might be configured to configure the space allocation of each block or page of memory cells independently from other blocks/pages of memory cells of the memory array 502. The control circuitry might comprise a memory device (e.g., non-volatile memory 516) in which memory array configuration data is stored. The configuration data might also be stored in a particular portion of the memory array 502, for example. During an initialization operation of the memory device (e.g., such as following a RESET operation), the stored configuration data might be read from the memory storing the configuration data and be loaded into the controller 506.
BLOCK10-BLOCK11 have been allocated by the control circuitry as ECC data storage space for the particular array configuration of
According to various embodiments, the size of each ECC data storage area (e.g., storage capacity) might be varied depending on a number of factors. According to one or more embodiments, the ECC data storage areas might be of the same size, such as shown by the storage areas storing ECC data ECC0-ECC7, for example. According to further embodiments, one or more of the ECC data storage areas might be of different sizes. The ECC data storage areas storing ECC data ECC0-ECC7 are shown to be a different size than the ECC data storage areas storing ECC data ECC8-ECC9. For example, a higher level of reliability might be desired for user data DATA8-DATA9 stored in BLOCK8-BLOCK9 of the memory array 602. Thus, a larger ECC data storage area might be allocated to store the ECC data needed to achieve the desired level of reliability of user data DATA8 and DATA9, for example. As the selection of ECC data storage area sizes might vary, spare storage area 608 in the ECC allocated portion 606 of the memory array might exist. However, a different configuration of the memory array according to one or more embodiments of the present disclosure might not result in the spare area 608 shown in
According to the configuration shown in
According to various embodiments, the memory array shown in
It should be noted that
Although the memory configuration and allocation with respect to
The memory device 1000 includes one or more memory arrays 1030 that might be logically arranged in banks of rows and columns. According to one or more embodiments, the memory cells of memory array 1030 are flash memory cells. The memory array 1030 might include multiple banks and blocks of memory cells residing on a single or multiple die as part of the memory device 1000. Memory array 1030 might comprise SLC and/or MLC memory. The memory array 1030 might also be adaptable to store varying densities (e.g., MLC (four level) and MLC (eight level)) of data in each cell, for example.
An address buffer circuit 1040 is provided to latch address signals provided on address input connections AO-Ax 1042. Address signals are received and decoded by a row decoder 1044 and a column decoder 1048 to access the memory array 1030. Row decoder 1044 might comprise driver circuits configured to drive the word lines of the memory array 1030, for example. It will be appreciated by those skilled in the art, with the benefit of the present description, that the number of address input connections 1042 might depend on the density and architecture of the memory array 1030. That is, the number of address digits increase with both increased memory cell counts and increased bank and block counts, for example.
The memory device 1000 reads data in the memory array 1030 by sensing voltage or current changes in the memory array columns using sense devices, such as sense/data cache circuitry 1050. The sense/data cache circuitry 1050, in at least one embodiment, is coupled to read and latch a row of data from the memory array 1030. Data input and output (I/O) buffer circuitry 1060 is included for bi-directional data communication over a plurality of data connections 1062 with the processor 1010. Write/erase circuitry 1056 is provided to write data to or to erase data from the memory array 1030.
Control circuitry 1070 is configured at least in part to implement various embodiments of the present disclosure, such as facilitating the memory configuration operations discussed above, for example. In at least one embodiment, the control circuitry 1070 may utilize a state machine. Control circuitry 1070 might be similar in configuration and functionality as control circuitry 506 discussed above with respect to
Control signals and commands can be sent by the processor 1010 to the memory device 1000 over the command bus 1072. The command bus 1072 may be a discrete signal or may be comprised of multiple signals, for example. These command signals 1072 are used to control the operations on the memory array 1030, including data read, data write (e.g., program), and erase operations. The command bus 1072, address bus 1042 and data bus 1062 may all be combined or may be combined in part to form a number of standard interfaces 1078. For example, the interface 1078 between the memory device 1000 and the processor 1010 might be a Universal Serial Bus (USB) interface. The interface 1078 might also be a standard interface used with many hard disk drives (e.g., SATA, PATA) as are known to those skilled in the art.
The electronic system illustrated in
In summary, one or more embodiments of the present disclosure provide methods of configuring user data and overhead data memory cells in a non-volatile memory device. These methods might facilitate a more efficient allocation of memory cells allocated during the configuration of a memory device for a particular application. These methods might facilitate a significant reduction in allocating more spare memory cells than might be needed given a particular application that the memory device is to be used in. More efficient memory device operation resulting from utilizing one or more of these methods of tailoring the allocation of memory cells in response to a particular memory device application might also be recognized.
Although specific embodiments have been illustrated and described herein, it will be appreciated by those of ordinary skill in the art that other configurations may be substituted for the specific embodiments shown. Many adaptations of the disclosure will be apparent to those of ordinary skill in the art. Accordingly, this application is intended to cover any adaptations or variations of the disclosure.
This application is a divisional of U.S. application Ser. No. 15/414,699, filed Jan. 25, 2017, and issued as U.S. Pat. No. 10,409,673 on Sep. 10, 2019, which is a divisional of U.S. application Ser. No. 14/513,880, filed Oct. 14, 2014 and issued as U.S. Pat. No. 9,594,676 on Mar. 14, 2017, which is a divisional of U.S. application Ser. No. 13/208,802, filed Aug. 12, 2011 and issued as U.S. Pat. No. 8,902,653 on Dec. 2, 2014, which are commonly assigned and incorporated in their entirety herein by reference.
Number | Name | Date | Kind |
---|---|---|---|
5986933 | Takeuchi et al. | Nov 1999 | A |
6961890 | Smith | Nov 2005 | B2 |
7155559 | Estakhri et al. | Dec 2006 | B1 |
7350044 | Keays | Mar 2008 | B2 |
7546515 | Gross et al. | Jun 2009 | B2 |
7702954 | Yoshida | Apr 2010 | B2 |
7870457 | Mokhlesi | Jan 2011 | B2 |
8064258 | Han | Nov 2011 | B2 |
8612836 | Yoon et al. | Dec 2013 | B2 |
8902653 | Radke et al. | Dec 2014 | B2 |
9594676 | Radke | Mar 2017 | B2 |
10409673 | Radke | Sep 2019 | B2 |
20040145952 | Chen et al. | Jul 2004 | A1 |
20050228962 | Takase et al. | Oct 2005 | A1 |
20050268203 | Keays | Dec 2005 | A1 |
20060206770 | Chen et al. | Sep 2006 | A1 |
20060248434 | Radke et al. | Nov 2006 | A1 |
20070220396 | Jackson | Sep 2007 | A1 |
20070234183 | Hwang et al. | Oct 2007 | A1 |
20080016428 | Lee et al. | Jan 2008 | A1 |
20080091971 | Yoshida | Apr 2008 | A1 |
20080098190 | Takase et al. | Apr 2008 | A1 |
20080282106 | Shalvi et al. | Nov 2008 | A1 |
20090150597 | Yang et al. | Jun 2009 | A1 |
20090259799 | Wong | Oct 2009 | A1 |
20090259807 | Estakhri et al. | Oct 2009 | A1 |
20100011276 | Chen et al. | Jan 2010 | A1 |
20100017682 | Jackson | Jan 2010 | A1 |
20100030945 | Yuan et al. | Feb 2010 | A1 |
20100199125 | Reche | Aug 2010 | A1 |
20100251066 | Radke | Sep 2010 | A1 |
20100262755 | Becker et al. | Oct 2010 | A1 |
20100281341 | Wu et al. | Nov 2010 | A1 |
20120030540 | Bacchin | Feb 2012 | A1 |
20130212449 | O'Connell | Aug 2013 | A1 |
Number | Date | Country |
---|---|---|
10-177797 | Jun 1998 | JP |
2003-131954 | May 2003 | JP |
2009-282923 | Dec 2009 | JP |
Entry |
---|
Chinese Office Action for CN Application No. 201280044579.0 dated Sep. 2, 2015 (6 Pages). The US counterpart of CN101079322A is 2007/234183. The US counterpart of CN1754230A is 2004/145952. |
European Search Report for EP Application No. 12823624.7 dated May 13, 2015 (6 pages). |
Number | Date | Country | |
---|---|---|---|
20190340065 A1 | Nov 2019 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 15414699 | Jan 2017 | US |
Child | 16516510 | US | |
Parent | 14513880 | Oct 2014 | US |
Child | 15414699 | US | |
Parent | 13208802 | Aug 2011 | US |
Child | 14513880 | US |