Embodiments of the disclosure relate generally to memory systems, and more specifically, relate to apparatuses and methods for logical block construction for physical blocks comprising multiple erase blocks coupled to a same string.
A memory system can include a memory sub-system, which can be a storage device, a memory module, or a hybrid of a storage device and a memory module. Examples of a storage device include a solid-state drive (SSD), a Universal Flash Storage (UFS) drive, a secure digital (SD) card, an embedded Multiple Media Card (eMMC), and a hard disk drive (HDD). Examples of memory modules include a dual in-line memory module (DIMM) and various types of non-volatile dual in-line memory modules (NVDIMMs). Memory systems include one or more memory components (e.g., memory devices) that store data. The memory components can be, for example, non-volatile memory components (e.g., NAND flash memory devices) and volatile memory components (e.g., DRAM devices). In general, a host system can utilize a memory system to store data at the memory components and to retrieve data from the memory components.
The present disclosure will be understood more fully from the detailed description given below and from the accompanying drawings of various embodiments of the disclosure.
Aspects of the present disclosure are directed to apparatuses and methods for logical block construction for physical blocks comprising multiple erase blocks coupled to a same string. Various types of memory, such as NAND flash memory, include a memory array of many memory cells that can be arranged in row and column fashion and grouped in physical blocks. The cells can include a charge storage node such as a floating gate or charge-trap layer which allows the cells to be programmed to store one more bits by adjusting the charge stored on the storage node. Generally, an erase operation (e.g., a “block erase”) is performed to erase all of the cells of a physical block together as a group.
Three-dimensional (3D) flash memory (e.g., a 3D NAND memory array) can include multiple strings of memory cells with each string comprising multiple series-coupled (e.g., source to drain) memory cells in a vertical direction, with the memory cells of a string sharing a common channel region. Each memory cell of a string can correspond to a different tier of the memory array, with a group of strings sharing multiple access lines, which may be referred to as word lines (WLs). Each access line can be coupled to respective memory cells of each string in the group of strings (e.g., the memory cells of a particular tier of the memory array). Groups of strings are coupled to respective sense lines, which may be referred to as data lines or bit lines (BLs), of a group of sense lines. The cells of the strings can be positioned between a drain-side select gate (referred to as a select gate drain (SGD)) and a source-side select gate (referred to as select gate source (SGS)) used to control access to the strings. A 3D NAND array can be a replacement gate (RG) NAND array or a floating gate NAND array, for example.
A 3D memory array can comprise multiple physical blocks each comprising a plurality of memory pages (e.g., physical pages of cells than can store one or more logical pages of data). In various previous approaches, a block of memory cells corresponds to a smallest group of memory cells that can be erased. For example, in prior approaches it is not possible to erase some of the memory cells of a block while maintaining data in other memory cells of the block.
Some prior approaches that may provide an ability to erase some memory cells of a block while maintaining data in other memory cells of the block can suffer various drawbacks. For example, independently operating groups of cells within a physical block can result in various disturb (e.g., program disturb, read disturb, program verify disturb, erase disturb, etc.) to the other groups of cells within the physical block. Such disturb can result in threshold voltage (Vt) shifts of the victim cells, which can lead to increased bit error rates (BERs) and/or data loss, for example.
As described further herein, some memory arrays can include physical blocks having multiple erase blocks therein. As used herein, an “erase block” refers to a group of cells that are configured to be erased together as a group and that share a same string as one or more additional groups of cells (e.g., one or more additional erase blocks). An erase block may also be referred to as a “deck.” As such, a physical block of cells can include multiple decks each capable of undergoing program/erase (P/E) cycling irrespective of the other decks.
Providing multiple erase blocks per physical block can provide benefits such as providing the ability to implement a logical block size that is smaller than a physical block size. A reduced logical block size can, for example, reduce the overhead associated with providing “spare” blocks used for system maintenance and/or other purposes. Since such spare blocks are not usable by the customer, it is beneficial to reduce the percentage overhead attributed to them. However, implementing a logical block size at the erase block level can introduce some challenges associated with operating (e.g., programming and/or erasing) the array.
For example, an erase operation can involve gate induced drain leakage (GIDL) hole injection into 3D NAND strings from either a drain side or a source side of the strings. If a physical block includes intermediate decks (e.g., erase blocks that are not adjacent to either a source side or a drain side of a string), it can be difficult to pass the generated GIDL holes to the intermediate decks, especially if the edge decks (e.g., erase blocks adjacent to either a source side or a drain side of the string) are in a programmed state. As used herein, intermediate decks may be referred to as “middle” decks.
Programming physical blocks comprising more than two decks, such that there are one or more middle decks, can also introduce operation complexities. For example, a programming operation is generally performed on an erased physical block and the physical pages (e.g., rows) are programmed sequentially from one end of the string to the other end of the string (e.g., source to drain or drain to source) with a seed voltage being passed through the string from one end to the other. Accordingly, since multiple decks within a physical block can be separately and independently programmed, it can be particularly difficult to program middle decks. For instance, if the seed voltage is generated from a drain side of the string and the deck adjacent to the drain side is programmed (e.g., with user data), it can be more difficult to pass the seed through the drain side deck to the middle deck than if the drain side deck was erased. Similarly, if the seed voltage is generated from a source side of the string and the deck adjacent to the source side is programmed (e.g., with user data), it can be more difficult to pass the seed through the source side deck to the middle deck than if the source side deck was erased.
Various embodiments of the present disclosure address the above and other deficiencies by providing apparatuses and methods for memory management involving an improved logical block architecture and implementation for memory arrays having physical blocks comprising multiple decks coupled to a same string. Various embodiments enable a logical block size that is smaller than a physical block size while overcoming challenges associated with operating (e.g., programming and/or erasing) physical blocks that include middle erase blocks that are not adjacent to either a source end or a drain end of string (e.g., physical blocks having three or more decks). Embodiments can increase the quantity of logical blocks per plane as compared to instances in which the logical block size matches the physical block size, which can enable a reduction in the quantity of physical blocks per plane without significantly increasing a spare block penalty. As described further herein, some embodiments in which physical blocks comprise three decks involve implementing (e.g., forming) three logical blocks out of two physical blocks (e.g., each logical block corresponding to a grouping of two decks), which implies a logical block size that is ⅔ (0.66) times the physical block size. The logical blocks are implemented such that each has access to either a source side of a string or a drain side of the string. The constituent decks of a particular logical block may be physically located in a same physical block (e.g., such that they are coupled to a same string) or in different physical blocks.
The memory device 100 includes control circuitry 110, address circuitry 112, and input/output (I/O) circuitry 114 used to communicate with an external device via an interface 119. The interface 119 can include, for example, a bus used to transmit data, address, and control signals, among other signals between the memory device 100 and an external host device, which can include a controller (e.g., system controller such as controller 891 shown in
The control circuitry 110 can decode signals (e.g., commands) received via interface 119 and executed to control operations performed on the memory array 102. The operations can include data programming operations, which may be referred to as write operations, data read operations, which may be referred to as sensing operations (and can include program verify operations), data erase operations, etc. The control circuitry 110 can cause various groups of memory cells (e.g., pages, blocks, erase blocks, etc.) to be selected or deselected in association with performing memory operations on the array 102. The control circuitry 110 can comprise a state machine, a sequencer, and/or some other type of control circuitry, which may be implemented in the form of hardware, firmware, or software, or any combination thereof.
The I/O circuitry 114 is used for bi-directional communication of data between the memory array 102 and the external device via interface 119. The address circuitry 112, which can include a register, can latch address signals received thereto, and the address signals can be decoded by a row decoder 116 and a column decoder 117 to access the memory array 102. The memory device 100 includes read/write circuitry 118 used to read data from and write data to the memory array 102. As an example, the read/write circuitry can include various latch circuitry, drivers, sense amplifiers, buffers, etc. Data can be read from the memory array 102 by sensing voltage and/or current changes on bit lines of the memory array 102.
The memory array 202 comprises a number of access lines (word lines) 222-0 (WL0), 222-1 (WL1), 222-2 (WL2), and 222-3 (WL3) and a number of sense lines (bit lines) 220-0 (BL0), 220-1 (BL1), and 220-2 (BL2) coupled to multiple strings 225-0-0, 225-0-1, 225-0-2, 225-1-0, 225-1-1, 225-1-2, 225-2-0, 225-2-1, and 225-2-2. The word lines, bit lines, and strings are collectively referred to as word lines 222, bit lines 220, and strings 225, respectively. Although four word lines 222, three bit lines 220, and nine strings 225 are shown, embodiments are not so limited.
Each of the strings 225 comprises a number of memory cells (referred to collectively as memory cells 223) located between a select transistor 224 and a select transistor 228. For example, as shown in
The memory cells 223 of the strings 225 are stacked vertically such that they are located on distinct tiers/levels of the memory array 202. Each word line 222 can be commonly coupled to all the memory cells at a particular tier/level. For example, word line 222-0 can be coupled to (e.g., as the control gate) the nine memory cells 223-0 corresponding to the nine respective strings 225.
The select gate transistors 224 and 228 can be controlled (e.g., turned on/off) via the corresponding select gate signals SGD0, SGD1, SGD2, SGS0, SGS1, and SGS2 in order to couple the strings 225 to their respective bit lines 220 and a common source line (SL) 229 during memory operations (e.g., reads, writes, erases). As shown in
To perform memory operations on the array 202, particular voltages (e.g., bias voltages) can be applied to the word lines 222, bit lines 220, and source line 229. The particular voltages applied depends on the memory operation being performed, and different voltages may be applied to the word lines 222 during a particular memory operation in order to store data in a cell (or page of cells) or read data from a cell. For example, an erase operation to remove data from a selected group of memory cells (e.g., a selected programmed deck) can include applying a relatively high voltage (e.g., 20V) to the source line 229, the relatively high voltage (e.g., 20V) to unselected word lines (e.g., word lines coupled to cells of a programmed deck and/or a deck not being erased), and a relatively low voltage (e.g., 0V) to the selected word lines (e.g., the word lines coupled to the deck being erased), which results in erasing of the cells of the selected programmed deck by removing charge from their charge storage nodes (e.g., charge-trap layers or floating gates) and thereby reducing their Vt levels to near 0V, for example. Additional example biasing schemes are described in more detail in association with
As described further in
As further described herein, an array (e.g., 202) can comprise a number of word lines physically between (e.g., separating) the word lines (e.g., 222) corresponding to different erase blocks. The word lines separating word lines corresponding to different erase blocks can be referred to as “dummy” word lines and can be coupled to dummy memory cells (e.g., within the strings 225) that are not used to store data. The dummy word lines and/or dummy cells can facilitate the ability to perform erase operations separately on erase blocks that share a common string or strings. For example, one erase block within a physical block can be erased without erasing one or more other erase blocks within the physical block. The quantity of dummy word lines between erase blocks can vary, and various bias voltages can be applied to the dummy word lines during the various memory operations performed on the erase blocks.
In operation, erase blocks can be separately (e.g., individually) selected or deselected. For example, an erase operation can be performed on a selected first erase block corresponding to a group of strings while another erase block(s) corresponding to the same group of strings is deselected for the erase operation (e.g., such that is not erased). As described further herein, the ability to independently operate (e.g., write, read, erase) erase blocks within a physical block can result in challenges associated with programming and/or erasing erase blocks that are not adjacent to a drain side or a source side of a string. Various embodiments described herein can address such challenges by implementing a logical block construction that involves grouping erased erase blocks in a manner such that middle erase blocks have access to either a source side or a drain side of a string.
In this example, the array 302 includes a plurality/group of word lines 322-1T, 322-2T, . . . , 322-NT corresponding to a first erase block 305-1 (e.g., a top deck) and a plurality/group of word lines 322-1B, 322-2B, . . . , 322-MB corresponding to a second erase block 305-2 (e.g., bottom deck). The designators “N” and “M” can represent various numbers (e.g., 3 or more) and “N” and “M” can be the same number. Accordingly, embodiments are not limited to a particular quantity of word lines 322 for the top deck 305-1 or bottom deck 305-2 (the designator “T” corresponding to “top” and the designator “B” corresponding to “bottom”). The array 302 also includes a number of dummy word lines 331-1, 331-2, 331-3, and 331-4, which can be collectively referred to as word lines 331. The dummy word lines 331 correspond to a separation region 333 between the top deck 305-1 and bottom deck 305-2. Although four word lines 331 are illustrated, embodiments can include more or fewer than four dummy word lines 331 separating erase blocks corresponding to same strings.
The array portion 302 illustrates two strings 325-1 and 325-2 for ease of illustration; however, embodiments can include many more strings 325. Memory cells are located at the intersections of the word lines 322/331 and strings 325, with the memory cells of a particular string 325 sharing a common channel region (e.g., pillar) as described in
As illustrated in
As noted herein, in various embodiments, the top deck 305-1 and the bottom deck 305-2 can be read, programmed, and/or erased via separate operations even though the cells of the decks 305-1/305-2 share the same strings 325-1/325-2. For example, each one of the decks 305-1 and 305-2 can be individually programmed and/or erased without programming or erasing the other of the decks 305-1 and 305-2.
Although only two decks (e.g., a top deck and a bottom deck) are illustrated in
Each of the physical blocks 404-1, . . . , 404-B includes multiple erase blocks. In this example, the physical blocks 404 have different respective quantities of erase blocks; however, embodiments are not so limited (e.g., the physical blocks 404 can all have a same quantity of erase blocks). In this example, physical block 404-1 includes a first erase block 405-1 (DECK_1), a second erase block 405-2 (DECK_2), and a third erase block 405-3 (DECK_3) separated by respective regions 411, which can correspond to a region of dummy word lines such as word lines 331 shown in
As described above, the decks 405 within a physical block 404 are commonly coupled to the strings of the respective blocks 404 with the decks 405 being separately erasable via a block erase operation (e.g., deck 405-1 can be erased without erasing decks 405-2 and/or deck 405-3) and separately programmable. Although the physical blocks 404 are shown as including two decks or three decks, embodiments are not so limited. For example, the physical blocks 404 can include more than three.
Each deck 405 can comprise a number of physical pages, which can correspond to a “row” of the array corresponding to a particular word line. As shown, deck 405-1 comprises pages 406-1-1, 406-1-2, . . . , 406-1-P, deck 405-2 comprises pages 406-2-1, 406-2-2, . . . , 406-2-P, and deck 405-3 comprises pages 406-3-1, 406-3-2, . . . , 406-3-P. The designator “P” is used to indicate that the decks 405 can comprise a plurality of pages/rows. Each physical page (collectively referred to as pages 406) can store multiple logical pages of data. A page can refer to a unit of programming and/or reading (e.g., a group of cells that are programmed and/or read together as a functional group).
As described further below in association with
The examples of
The example of
In contrast, the example of
Providing the ability to dynamically form the logical blocks 540 can provide various benefits. For example, a particular combination of logical block candidates 540-1 to 540-6 can be selected based on an amount of valid data within the respective erase blocks 505. For instance, a combination can be selected that results in a least amount of valid data, which can reduce the garbage collection and/or write amplification associated with operating the array. The amount of valid data per erase block 505 can be tracked, for example, by a controller such as controller 891 described in
The example of
The physical blocks of the planes can comprise multiple erase blocks sharing common strings as described herein. The physical blocks can be grouped into “super blocks” with each super block comprising a physical block from each plane (e.g., PLANE 0 and PLANE 1) across multiple LUNs (e.g., across multiple arrays 702). Similarly, embodiments of the present disclosure an include a number of super decks 715-1 (SUPER DECK_1), 715-2 (SUPER DECK_2), . . . , 715-D (SUPER DECK_D). Each super deck (or super erase block) 715 can comprise a deck from each plane across multiple LUNs. For example, a first super deck 715-1 (SUPER DECK_1) can comprise a deck from plane 0 of LUN0, a deck from plane 1 of LUN1, a deck from plane 0 of LUN1, a deck from plane 1 of LUN1, a deck from plane 0 of LUN2, a deck from plane 1 of LUN2, a deck from plane 0 of LUN3, and a deck from plane 1 of LUN3.
Embodiments of the present disclosure can monitor temperature classification information on a super deck level as well as, or instead of, on a deck level. For instance, consider an example in which the constituent decks of a super deck 715-1 share common strings with the respective constituent decks of a super deck 715-2 (e.g., super decks 715-1 and 715-2 are located in a same physical super block). The decks of super deck 715-1 can be erased together as a group.
In some embodiments, the memory system 890 is a storage system. An example of a storage system is a solid-state drive (SSD). In some embodiments, the memory system 890 is a hybrid memory/storage sub-system. In general, the computing environment shown in
The memory system controller 891 (hereinafter referred to as “controller”) can communicate with the memory devices 800 to perform operations such as reading data, writing data, or erasing data at the memory devices 800 and other such operations. The controller 891 can include hardware such as one or more integrated circuits and/or discrete components, a buffer memory, or a combination thereof. The controller 891 can include a microcontroller, special purpose logic circuitry (e.g., a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), etc.), or other suitable processing circuitry. The controller 891 can include a processing device (e.g., processor 894) configured to execute instructions stored in local memory (not shown).
In this example, the controller 891 includes a memory management component 813. The memory management component 813 can be associated with a flash translation layer (FTL), for example. The component 813 can be responsible for, among other things, operating a memory array of memory device 800 in accordance with a logical block implementation in which each logical block comprises: a first erase block adjacent to a first end of a particular string corresponding to a first physical block; and a second erase block. The second erase block is either: located in the first physical block and not adjacent to a second end of the particular string corresponding to the first physical block; or located in a second physical block and adjacent to a first end of a particular string corresponding to the second physical block. The memory management component can also be responsible for tracking an amount of valid data on a per erase block basis, and dynamically group logical blocks based, at least partially, on the amount of valid data per erase block.
In general, the controller 891 can receive commands or operations from the host system 892 and can convert the commands or operations into instructions or appropriate commands to achieve the desired access to the memory devices 800. The controller 891 can be responsible for other operations such as wear leveling operations, garbage collection operations, error detection and error-correcting code (ECC) operations, encryption operations, caching operations, and address translations between a logical block address and a physical block address that are associated with the memory devices 800. Garbage collection can involve moving valid data (e.g., pages) of an erase block (e.g., source block) to a different erase block (e.g., destination block) in order to erase the source block and add it to a pool of available free erase blocks. Garbage collection is often performed as a background operation (e.g., by system controller 891) such that it is transparent to the host 892.
The host system 892 can be a computing device such as a desktop computer, laptop computer, network server, mobile device, or other such computing device that includes a memory and a processing device. The host system 892 can include, or be coupled to, the memory system 890 so that the host system 892 can read data from or write data to the memory system 890. The host system 892 can be coupled to the memory system 890 via a physical host interface (not shown in
While the example memory system 890 in
Although the memory system 890 is shown as physically separate from the host 892, in a number of embodiments the memory system 890 can be embedded within the host 892. Alternatively, the memory system 890 can be removable from the host 892.
As used herein, an “apparatus” can refer to various structural components. For example, the computing system 801 shown in
Various methods of the present disclosure can be performed by processing logic that can include hardware (e.g., processing device, circuitry, dedicated logic, programmable logic, microcode, hardware of a device, integrated circuit, etc.), software (e.g., instructions run or executed on a processing device), or a combination thereof. In some embodiments, the method, or portions thereof are performed by the control circuitry 110 of
One example method can include forming, via a controller coupled to a memory array comprising a plurality of physical blocks of memory cell, logical block groupings for the erase blocks in accordance with a particular logical block implementation. Each physical block of the plurality can comprise more than two erase blocks, with each of the more than two erase blocks of each respective physical block comprising memory cells coupled to a same string of memory cells corresponding to the respective physical block. The logical block groupings can be formed such that each logical block comprises: a first erase block adjacent to a first end of a particular string corresponding to a first physical block; and a second erase block. The second erase block is either: located in the first physical block and not adjacent to a second end of the particular string corresponding to the first physical block; or located in a second physical block and adjacent to a first end of a particular string corresponding to the second physical block.
In a number of embodiments, the method can include determining the logical block groupings such that any erase blocks that are not adjacent to either end of the particular string corresponding to the first physical block or either end of the particular string corresponding to the second physical block are adjacent to an erase block that is adjacent to either end of the particular string corresponding to the first physical block or either end of the particular string corresponding to the second physical block. Various embodiments can include tracking, via the controller, an amount of valid data on an erase block basis, and dynamically adjusting, via the controller, the logical block groupings based on determined changes in the amount of valid data per erase block. The erase blocks can be grouped to facilitate passage of a seed voltage to intermediate erase blocks in association with programming the intermediate erase blocks. The example method can include performing, via the controller, an operation (e.g., a programming and/or erase operation) on the erase blocks based on the determined logical block groupings.
Some portions of the preceding detailed descriptions have been presented in terms of algorithms and symbolic representations of operations on data bits within a computer memory. These algorithmic descriptions and representations are the ways used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. An algorithm is here, and generally, conceived to be a self-consistent sequence of operations leading to a desired result. The operations are those requiring physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, combined, compared, and otherwise manipulated. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like.
It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. The present disclosure can refer to the action and processes of a computer system, or similar electronic computing device, which manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage systems.
The present disclosure also relates to an apparatus for performing the operations herein. This apparatus can be specially constructed for the intended purposes, or it can include a general purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program can be stored in a computer readable storage medium, such as, but not limited to, any type of disk including floppy disks, optical disks, CD-ROMs, and magnetic-optical disks, read-only memories (ROMs), random access memories (RAMs), EPROMs, EEPROMs, magnetic or optical cards, or any type of media suitable for storing electronic instructions, each coupled to a computer system bus.
The algorithms and displays presented herein are not inherently related to any particular computer or other apparatus. Various general purpose systems can be used with programs in accordance with the teachings herein, or it can prove convenient to construct a more specialized apparatus to perform the method. The structure for a variety of these systems will appear as set forth in the description below. In addition, the present disclosure is not described with reference to any particular programming language. It will be appreciated that a variety of programming languages can be used to implement the teachings of the disclosure as described herein.
The present disclosure can be provided as a computer program product, or software, which can include a machine-readable medium having stored thereon instructions, which can be used to program a computer system (or other electronic devices) to perform a process according to the present disclosure. A machine-readable medium includes any mechanism for storing information in a form readable by a machine (e.g., a computer). In some embodiments, a machine-readable (e.g., computer-readable) medium includes a machine (e.g., a computer) readable storage medium such as a read only memory (“ROM”), random access memory (“RAM”), magnetic disk storage media, optical storage media, flash memory components, etc.
The figures herein follow a numbering convention in which the first digit or digits correspond to the drawing figure number and the remaining digits identify an element or component in the drawing. Similar elements or components between different figures may be identified by the use of similar digits. For example, 102 may reference element “02” in
For the purposes of the present disclosure, the phrase “A and/or B” means (A), (B), (A) or (B), or (A and B). For the purposes of the present disclosure, the phrase “A, B, and/or C” means (A), (B), (C), (A and B), (A and C), (B and C), or (A, B and C). Additionally, the phrase “at least one of A and B” means one or more of (A) or one or more of (B), or one or more of (A) and one or more of (B) such that both one or more of (A) and one or more of (B) is not required.
In the foregoing specification, embodiments of the disclosure have been described with reference to specific example embodiments thereof. It will be evident that various modifications can be made thereto without departing from the broader spirit and scope of embodiments of the disclosure as set forth in the following claims. The specification and drawings are, accordingly, to be regarded in an illustrative sense rather than a restrictive sense.
This application claims the benefits of U.S. Provisional Application No. 63/545,581, filed on Oct. 25, 2023, the contents of which are incorporated herein by reference.
Number | Date | Country | |
---|---|---|---|
63545581 | Oct 2023 | US |