Storage device

The storage device addresses inefficiencies in data classification by using a controller to determine cold data based on reclaim counts, enhancing performance and reliability through optimized data management in non-volatile memory systems.

KR103045533B1Active Publication Date: 2026-11-06SAMSUNG ELECTRONICS CO LTD
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
KR1020210133845
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-10-08
Publication Date
2026-11-06
Estimated Expiration
2041-10-08

AI Technical Summary

Technical Problem

Existing storage devices struggle with inefficient data classification and management, particularly in non-volatile memory systems, leading to performance and reliability issues due to the lack of effective methods for identifying and handling cold data.

Method used

A storage device with a non-volatile memory system that includes a controller for performing patrol reads and increasing a reclaim count for user data, determining data as cold based on the reclaim count exceeding a threshold, and transferring such data to a different memory block for improved management.

Benefits of technology

Enhances performance and reliability by effectively classifying and managing cold data using reclaim counts, optimizing data distribution across memory blocks, thereby improving wear-leveling and garbage collection processes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112021115625535-PAT00008_ABST
    Figure 112021115625535-PAT00008_ABST
Patent Text Reader

Abstract

A storage device with improved performance and reliability is provided. The storage device includes a non-volatile memory comprising a first memory block for storing first user data, and a controller that performs a patrol read on the first user data. The controller increases a reclaim count for the first user data when a reclaim for the first user data is required, and when the reclaim count for the first user data is greater than a reclaim threshold, determines the first user data as cold data, performs a first reclaim on the first user data determined as cold data, and the non-volatile memory stores the first user data in a second memory block different from the first memory block in response to the first reclaim.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a storage device. Background Technology

[0002] Semiconductor memory devices include volatile memory devices and non-volatile memory devices. While volatile memory devices have fast read and write speeds, they can lose stored content when the power is turned off. Conversely, since non-volatile memory devices retain stored content even when the power is off, they are used to store content that must be maintained regardless of power supply.

[0003] For example, volatile memory devices include static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), etc. Non-volatile memory devices retain stored contents even when the power is off. For example, non-volatile memory devices include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically eraseable and programmable ROM (EEPROM), flash memory, phase change RAM (PRAM), magnetic RAM (MRAM), resistive RAM (RRAM), ferroelectric RAM (FRAM), etc. Flash memory can be classified into NOR-type flash memory and NAND-type flash memory.

[0004] Referring to Korean Published Patent No. 2021-0003625, when performing read reclaim, the central processing unit can distinguish data stored in a non-super block into hot data and cold data, and copy the hot data to the super block. Here, hot data refers to data with a high number of reads, and cold data refers to data with a low number of reads. That is, according to the prior art, the storage device classifies data as cold data based on the number of reads. The method of distinguishing cold data by the storage device may be problematic. The problem to be solved

[0005] The technical problem that the present invention aims to solve is to provide a storage device with improved performance and reliability.

[0006] The technical problems of the present invention are not limited to those mentioned above, and other unmentioned technical problems will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0007] A storage device according to some embodiments for achieving the above technical problem includes a non-volatile memory comprising a first memory block for storing first user data, and a controller for performing a patrol read on the first user data. The controller increases a reclaim count for the first user data when a reclaim for the first user data is required, and when the reclaim count for the first user data is greater than a reclaim threshold, determines the first user data as cold data and performs a first reclaim on the first user data determined as cold data. The non-volatile memory may store the first user data in a second memory block different from the first memory block in response to the first reclaim.

[0008] A storage device according to some embodiments for achieving the above technical problem includes a non-volatile memory and a controller that receives first user data, a logical block address, and a write command from an external device, creates a first matching table for a first address that matches the logical block address, and provides the first address and the first user data to the non-volatile memory, wherein the non-volatile memory stores the first user data based on the first address, and the controller increases a reclaim count for the first user data, performs a reclaim for the first user data, creates a second matching table for a second address that matches the logical block address, and the second address may be different from the first address.

[0009] A storage device according to some embodiments for achieving the above technical problem includes a first non-volatile memory for storing first user data, a second non-volatile memory different from the first non-volatile memory, and a controller that performs a patrol read on the first user data, wherein the controller increases a reclaim count for the first user data when reclaiming the first user data is required, and when the reclaim count for the first user data is greater than a reclaim threshold value, provides the first user data to the second non-volatile memory, and the second non-volatile memory can store the first user data.

[0010] Specific details of other embodiments are included in the detailed description and drawings. Brief explanation of the drawing

[0011] FIG. 1 is a block diagram of a memory system according to some embodiments. Figure 2 is a block diagram of the non-volatile memory of Figure 1. Figure 3 is a block diagram of the storage controller and non-volatile memory of Figure 1. FIG. 4 is an exemplary circuit diagram illustrating a memory cell array according to some embodiments. FIG. 5 is a block diagram of a reclaim controller and non-volatile memory according to some embodiments. FIG. 6 is a ladder diagram for explaining the operation of a memory system according to some embodiments. Figure 7 is a diagram illustrating a patrol lead. FIG. 8 is a flowchart illustrating reclaim count-up and cold data reclaim according to some embodiments. Figure 9 is a block diagram of a non-volatile memory where reclaiming has been performed. Figure 10 is a flowchart explaining the operation of the threshold value determiner of Figure 5. FIG. 11 is a diagram showing a change in a matching table according to some embodiments. FIG. 12 is a ladder diagram for explaining the read operation of first user data according to some embodiments. FIGS. 13 and FIGS. 14 are block diagrams of a storage device including a plurality of memory blocks according to some embodiments. FIG. 15 relates to a memory system comprising a plurality of storage devices according to some embodiments. Figure 16 is a ladder diagram to explain the operation of the memory system of Figure 15. FIG. 17 is a drawing of a data center including a storage device according to some embodiments. Specific details for implementing the invention

[0012] Embodiments according to the technical concept of the present invention will be described below with reference to the attached drawings.

[0013] FIG. 1 is a block diagram of a memory system according to some embodiments.

[0014] Referring to FIG. 1, the memory system (1) may include a host device (20) and a storage device (10). The host device (20) may include a host controller (21) and a host memory (22). The host controller (21) may control the overall operation of the host device (20). The host memory (22) may temporarily store data transmitted from the outside, data to be transmitted to the storage device (10), or data transmitted from the storage device (10). The host device (20) may be implemented as an application processor (AP). However, embodiments of the present invention are not limited thereto.

[0015] The storage device (10) may include a storage controller (200) and a non-volatile memory (400).

[0016] A storage device (10) may include storage media for storing data in response to a request from a host device (20). As an example, the storage device (10) may include at least one of a Solid State Drive (SSD), embedded memory, and removable external memory. If the storage device (10) is an SSD, the storage device (10) may be a device that follows the NVMe (non-volatile memory express) standard. If the storage device (10) is embedded memory or external memory, the storage device (10) may be a device that follows the UFS (universal flash storage) or eMMC (embedded multi-media card) standard. The storage device (10) and the host device (20) may each generate and transmit packets according to the adopted standard protocol.

[0017] When the non-volatile memory (400) of the storage device (10) includes flash memory, the flash memory may include a 2D NAND memory array or a 3D (or vertical) NAND (VNAND) memory array. As another example, the storage device (10) may include various other types of non-volatile memory. For example, the storage device (10) may include Magnetic RAM (MRAM), Spin-Transfer Torque MRAM, Conductive Bridging RAM (CBRAM), Ferroelectric RAM (FeRAM), Phase RAM (PRAM), Resistive RAM, and various other types of memory.

[0018] The storage controller (200) may include a host interface (211), a memory interface (212), and a central processing unit (CPU) (213). Additionally, the storage controller (210) may further include a Flash Translation Layer (FTL) (214), a packet manager (215), a buffer memory (216), an error correction code (ECC) (217) engine, and an advanced encryption standard (AES) engine (218). The storage controller (200) may further include a working memory (not shown) on which the Flash Translation Layer (FTL) (214) is loaded, and data write and read operations for non-volatile memory (400) may be controlled by the CPU (213) executing the Flash Translation Layer.

[0019] The host interface (211) can transmit and receive packets with the host device (20). The packet transmitted from the host device (20) to the host interface (211) may include a command or data to be written to the non-volatile memory (400), and the packet transmitted from the host interface (211) to the host device (20) may include a response to the command or data read from the non-volatile memory (400). The memory interface (212) can transmit data to be written to the non-volatile memory (400) to the non-volatile memory (400) or receive data read from the non-volatile memory (400). This memory interface (212) may be implemented to comply with standard protocols such as Toggle or Open NAND Flash Interface (ONFI).

[0020] The flash conversion layer (214) can perform various functions such as address mapping, wear-leveling, and garbage collection. The address mapping operation is an operation that converts a logical address received from the host device (20) into a physical address used to actually store data in the non-volatile memory (400). For example, the storage controller (200) can create a matching table containing a logical block address and a corresponding physical block address. Wear-leveling is a technique to prevent excessive degradation of a specific block by ensuring that blocks in the non-volatile memory (400) are used evenly, and can be implemented, for example, through firmware technology that balances the erase counts of physical blocks. Garbage collection is a technique to secure usable capacity in the non-volatile memory (400) by erasing the existing block after copying the valid data of the block to a new block. In some embodiments, the storage controller (200) can perform functions such as address mapping, wear-leveling, and garbage collection by referring to metadata stored in non-volatile memory (400).

[0021] The packet manager (215) can generate a packet according to the protocol of the interface agreed upon with the host device (20) or parse various information from a packet received from the host device (20).

[0022] The buffer memory (216) can temporarily store data to be written to the non-volatile memory (400) or data to be read from the non-volatile memory (400). The buffer memory (216) may be configured to be located within the storage controller (200), but it may also be placed outside the storage controller (200).

[0023] The ECC engine (217) can perform error detection and correction functions for data read from the non-volatile memory (400). Specifically, the ECC engine (217) can generate parity bits for data to be written to the non-volatile memory (400), and the generated parity bits can be stored in the non-volatile memory (400) together with the written data. When reading data from the non-volatile memory (400), the ECC engine (217) can correct errors in the read data using the parity bits read from the non-volatile memory (400) together with the read data, and output the read data with the errors corrected.

[0024] The AES engine (218) can perform at least one of the encryption and decryption operations on data input to the storage controller (200) using a symmetric-key algorithm.

[0025] Figure 2 is a block diagram of the non-volatile memory of Figure 1.

[0026] Referring to FIG. 2, the non-volatile memory (400) may include a memory cell array (410), an address decoder (420), a voltage generator (430), a read-write circuit (440), and a control logic circuit (450).

[0027] The memory cell array (410) can be connected to an address decoder (420) via word lines (WL). The memory cell array (410) can be connected to a read-write circuit (440) via bit lines (BL). The memory cell array (410) may include a plurality of memory cells. For example, memory cells arranged in a row direction can be connected to word lines (WL). For example, memory cells arranged in a column direction can be connected to bit lines (BL).

[0028] The address decoder (420) can be connected to the memory cell array (410) via a word line (WL). The address decoder (420) can operate in response to the control of the control logic circuit (450). The address decoder (420) can receive an address (ADDR) from the storage controller (200). The address decoder (420) can receive a voltage required for operations such as programming and reading from the voltage generator (430).

[0029] The address decoder (420) can decode the row address among the received addresses (ADDR). The address decoder (420) can select a word line (WL) using the decoded row address. The decoded column address (DCA) can be provided to the read-write circuit (440). For example, the address decoder (420) may include a row decoder, a column decoder, and an address buffer, etc.

[0030] The voltage generator (430) can generate a voltage required for an access operation under the control of the control logic circuit (450). For example, the voltage generator (430) can generate a program voltage and a program verification voltage required to perform a program operation. For example, the voltage generator (430) can generate read voltages required to perform a read operation, and generate an erase voltage and an erase verification voltage required to perform an erase operation. For example, the voltage generator (430) can generate a monitoring voltage for monitoring data stored in the memory cell array (410). Additionally, the voltage generator (430) can provide the voltage required to perform each operation to the address decoder (420).

[0031] The read-write circuit (440) can be connected to the memory cell array (410) via a bit line (BL). The read-write circuit (440) can exchange data (DATA) with the storage controller (200). The read-write circuit (440) can operate in response to the control of the control logic circuit (450). The read-write circuit (440) can receive a decoded column address (DCA) from the address decoder (420). The read-write circuit (440) can select a bit line (BL) using the decoded column address (DCA).

[0032] For example, the read-write circuit (440) can program received data (DATA) into the memory cell array (410). The read-write circuit (440) can read data from the memory cell array (410) and provide the read data to an external source (e.g., a storage controller (200)). For example, the read-write circuit (440) may include a configuration such as a sense amplifier, a write driver, a column selector circuit, and a page buffer. That is, the read-write circuit (440) can buffer the data (DATA) received from the storage controller (200) into a page buffer and program the buffered data (DATA) into the memory cell array (410).

[0033] The control logic circuit (450) can be connected to an address decoder (420), a voltage generator (430), and a read / write circuit (440). The control logic circuit (450) can control the operation of the non-volatile memory (400). The control logic circuit (450) can operate in response to a control signal (CRTL) and a command (CMD) (e.g., a write command and a read command, etc.) provided by the storage controller (200).

[0034] Figure 3 is a block diagram of the storage controller and non-volatile memory of Figure 1.

[0035] Referring to FIG. 3, the storage device (10) may include a storage controller (200) and a non-volatile memory (400). The storage device (10) may support a plurality of channels (CH1 to CHm), and the storage controller (200) and the non-volatile memory (400) may be connected through the plurality of channels (CH1 to CHm). For example, the storage device (10) may be implemented as a storage device such as a Solid State Drive (SSD).

[0036] The non-volatile memory (400) may include a plurality of non-volatile memory devices (NVM11 to NVMmn). Each of the non-volatile memory devices (NVM11 to NVMmn) may be connected to one of a plurality of channels (CH1 to CHm) through a corresponding way. For example, the non-volatile memory devices (NVM11 to NVM1n) may be connected to a first channel (CH1) through ways (W11 to W1n), and the non-volatile memory devices (NVM21 to NVM2n) may be connected to a second channel (CH2) through ways (W21 to W2n). In an exemplary embodiment, each of the non-volatile memory devices (NVM11 to NVMmn) may be implemented as any memory unit capable of operating according to individual commands from the storage controller (200). For example, each of the non-volatile memory devices (NVM11 to NVMmn) may be implemented as a chip or a die, but the present invention is not limited thereto.

[0037] The storage controller (200) can transmit and receive signals to and from the non-volatile memory (400) through multiple channels (CH1~CHm). For example, the storage controller (200) can transmit commands (CMDa~CMDm), addresses (ADDRa~ADDRm), and data (DATAa~DATAm) to the non-volatile memory (400) or receive data (DATAa~DATAm) from the non-volatile memory (400) through the channels (CH1~CHm).

[0038] The storage controller (200) can select one of the non-volatile memory devices connected to the respective channel through each channel and transmit and receive signals with the selected non-volatile memory device. For example, the storage controller (200) can select a non-volatile memory device (NVM11) among the non-volatile memory devices (NVM11 to NVM1n) connected to the first channel (CH1). The storage controller (200) can transmit a command (CMDa), an address (ADDRa), and data (DATAa) to the selected non-volatile memory device (NVM11) through the first channel (CH1), or receive data (DATAa) from the selected non-volatile memory device (NVM11).

[0039] The storage controller (200) can transmit and receive signals to and from the non-volatile memory (400) in parallel through different channels. For example, the storage controller (200) can transmit a command (CMDb) to the non-volatile memory (400) through the second channel (CH2) while transmitting a command (CMDa) to the non-volatile memory (400) through the first channel (CH1). For example, the storage controller (200) can receive data (DATAb) from the non-volatile memory (400) through the second channel (CH2) while receiving data (DATAa) from the non-volatile memory (400) through the first channel (CH1).

[0040] The storage controller (200) can control the overall operation of the non-volatile memory (400). The storage controller (200) can control each of the non-volatile memory devices (NVM11~NVMmn) connected to the channels (CH1~CHm) by transmitting signals to the channels (CH1~CHm). For example, the storage controller (200) can control one of the selected non-volatile memory devices (NVM11~NVM1n) by transmitting a command (CMDa) and an address (ADDRa) to the first channel (CH1).

[0041] Each of the non-volatile memory devices (NVM11 to NVMmn) can operate under the control of the storage controller (200). For example, the non-volatile memory device (NVM11) can program data (DATAa) according to a command (CMDa), an address (ADDRa), and data (DATAa) provided through the first channel (CH1). For example, the non-volatile memory device (NVM21) can read data (DATAb) according to a command (CMDb) and an address (ADDRb) provided through the second channel (CH2), and transmit the read data (DATAb) to the storage controller (200).

[0042] FIG. 3 illustrates that a non-volatile memory (400) communicates with a storage controller (200) through m channels, and that the non-volatile memory (400) includes n non-volatile memory devices corresponding to each channel, but the number of channels and the number of non-volatile memory devices connected to one channel can be varied.

[0043] FIG. 4 is an exemplary circuit diagram illustrating a memory cell array according to some embodiments.

[0044] Referring to FIG. 4, a plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33) may be arranged on a substrate (not shown) in a first direction (x) and a second direction (y). The plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33) may have a shape extended in a third direction (z). The plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33) may be commonly connected to a common source line (CSL) formed on or within the substrate (not shown). Although it is illustrated that a common source line (CSL) is connected to the bottom end of the third direction (z) of the plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33), it is sufficient for the common source line (CSL) to be electrically connected to the bottom end of the third direction (z) of the plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33), and is not limited to being physically located at the bottom end of the plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33). Additionally, although the plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33) are shown in the drawing as being arranged in a 3 x 3 array, the arrangement and number of the plurality of cell strings arranged in the memory cell array (410) are not limited thereto.

[0045] Some cell strings (NS11, NS12, and NS13) may be connected to the first ground select line (GSL: Ground Select Line) (GSL1). Some cell strings (NS21, NS22, and NS23) may be connected to the second ground select line (GSL2). Some cell strings (NS31, NS32, and NS33) may be connected to the third ground select line (GSL3).

[0046] Additionally, some cell strings (NS11, NS12, and NS13) may be connected to a first string select line (SSL: String Select Line) (SSL1). Some cell strings (NS21, NS22, and NS23) may be connected to a second string select line (SSL2). Some cell strings (NS31, NS32, and NS33) may be connected to a third string select line (SSL3).

[0047] Each of the plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33) may include a String Select Transistor (SST) connected to each string select line. Additionally, each of the plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33) may include a Ground Select Transistor (GST) connected to each ground select line.

[0048] One end of the ground select transistor of each of the plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33) may be connected to a common source line (CSL). Additionally, for each of the plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33), a plurality of memory cells may be stacked sequentially in a third direction (z) between the ground select transistor and the string select transistor. Although not shown in the drawing, each of the plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33) may include dummy cells between the ground select transistor and the string select transistor. Furthermore, the number of string select transistors included in each string is not limited to the drawing.

[0049] For example, the cell string (NS11) may include a ground select transistor (GST11) positioned at the bottom in the third direction (z), a plurality of memory cells (M11_1 to M11_8) stacked sequentially in the third direction (z) on the ground select transistor (GST11), and a string select transistor (SST11) stacked in the third direction (z) on the top memory cell (M11_8). Additionally, the cell string (NS21) may include a ground select transistor (GST21) positioned at the bottom in the third direction (z), a plurality of memory cells (M21_1 to M21_8) stacked sequentially in the third direction (z) on the ground select transistor (GST21), and a string select transistor (SST21) stacked in the third direction (z) on the top memory cell (M21_8). Additionally, the cell string (NS31) may include a ground select transistor (GST31) positioned at the bottom in the third direction (z), a plurality of memory cells (M31_1 to M31_8) stacked sequentially in the third direction (z) on the ground select transistor (GST31), and a string select transistor (SST31) stacked in the third direction (z) on the top memory cell (M31_8). Below, the configuration of other strings may be similar to this.

[0050] Memory cells located at the same height in the third direction (z) from a substrate (not shown) or a ground select transistor may be electrically connected in common through their respective word lines. For example, memory cells at a height where memory cells (M11_1, M21_1, and M31_1) are formed may be connected to the first word line (WL1). Additionally, memory cells at a height where memory cells (M11_2, M21_2, and M31_2) are formed may be connected to the second word line (WL2). The arrangement and structure of memory cells connected to the third word line (WL3) through the eighth word line (WL8) are similar thereto, so a description is omitted below.

[0051] One end of each string select transistor of a plurality of cell strings (NS11, NS21, NS31, NS12, NS22, NS32, NS13, NS23, and NS33) can be connected to a bit line (BL1, BL2, and BL3). For example, string select transistors (ST11, SST21, and SST31) can be connected to a bit line (BL1) extending in the second direction (y). The description of other string select transistors connected to the bit line (BL2, and BL3) is similar, so the description is omitted.

[0052] Memory cells corresponding to one string (or ground) select line and one word line can form one page. Write and read operations can be performed on a unit of each page. Each memory cell of each page may store two or more bits. The bits written to the memory cells of each page can form logical pages.

[0053] The memory cell array (410) may be provided as a three-dimensional memory array. The three-dimensional memory array may be monolithically formed on one or more physical levels of arrays of memory cells having active regions disposed on a substrate (not shown) and a circuit associated with the operation of the memory cells. The circuit associated with the operation of the memory cells may be located within or on the substrate. Monolithic formation means that the layers of each level of the three-dimensional array may be directly deposited on the layers of the lower levels of the three-dimensional array.

[0054] Hereinafter, a storage device that performs a reclaim operation will be described with reference to FIGS. 5 to 11.

[0055] FIG. 5 is a block diagram of a reclaim controller and non-volatile memory according to some embodiments. FIG. 6 is a ladder diagram for explaining the operation of a memory system according to some embodiments.

[0056] Referring to FIG. 5, the storage device (10) may include a reclaim controller (100) and a non-volatile memory (400). The reclaim controller (100) may be included in the storage controller (200) illustrated in FIG. 1. For example, the reclaim controller (100) may be implemented by the configuration of the storage controller (200). However, embodiments of the present invention are not limited thereto. For example, the reclaim controller (100) may be implemented by the non-volatile memory (400) and may also be implemented by firmware. In this specification, the reclaim controller (100) is described as controlling a reclaim operation for the non-volatile memory (400).

[0057] First, the non-volatile memory (400) may include a first memory block (410a) and a second memory block (410b). Here, the first memory block (410a) and the second memory block (410b) may be included in a memory cell array (410). That is, a part of the memory cell array (410) may correspond to the first memory block (410a), and another part of the memory cell array (410) may correspond to the second memory block (410b). The first memory block (410a) and the second memory block (410b) may be physically separated.

[0058] Additionally, the first memory block (410a) and the second memory block (410b) may have different characteristics. For example, the first memory block (410a) may have a memory cell corresponding to SLC (single level cell), and the second memory block (410b) may have a memory cell corresponding to at least one of MLC (multi level cell), TLC (triple level cell), and QLC (quad level cell).

[0059] Here, a memory cell corresponding to SLC can store data having one bit. However, a memory cell corresponding to MLC can store data having two bits, a memory cell corresponding to TLC can store data having three bits, and a memory cell corresponding to QLC can store data having four bits. That is, the capacity of data stored in a memory cell corresponding to SLC may be smaller than the capacity of data stored in memory cells corresponding to MLC, TLC, and QLC. Therefore, when the first memory block (410a) and the second memory block (410b) occupy the same area, the capacity of data stored in the first memory block (410a) may be smaller than the capacity of data stored in the second memory block (410b).

[0060] Additionally, the first memory block (410a) may include MLC, and the second memory block (410b) may include at least one of TLC and QLC. That is, the capacity of data stored in the memory cells of the first memory block (410a) may be smaller than the capacity of data stored in the memory cells of the second memory block (410b). Additionally, the operating speed of the memory cells of the first memory block (410a) may be greater than the operating speed of the memory cells of the second memory block (410b).

[0061] The reclaim controller (100) may include a patrol lead controller (110), a threshold determiner (120), a cold data determiner (130), and a cold data reclaim controller (140). The patrol lead controller (110), threshold determiner (120), cold data determiner (130), and cold data reclaim controller (140) illustrated in FIG. 5 are examples and do not physically exist. That is, the patrol lead controller (110), threshold determiner (120), cold data determiner (130), and cold data reclaim controller (140) may correspond to functions implemented by the reclaim controller (100) or the storage controller (200).

[0062] In some embodiments, the patrol read controller (110) may perform a patrol read on the non-volatile memory (400). The memory cell array (410) included in the non-volatile memory (400) may include a plurality of memory cells. Over time, the charge stored in the memory cells may change, and accordingly, the memory cells may degrade. For example, retention may occur in the non-volatile memory (400). That is, the dispersion of the memory cells may differ from the initial dispersion. Therefore, additional processing may be required for the retained memory cells.

[0063] The patrol read controller (110) can apply a monitoring voltage to the memory cells of the non-volatile memory (400). The patrol read controller (110) can determine whether the memory cells included in the non-volatile memory (400) are degraded. After data is programmed into the non-volatile memory (400) and after a long period of time has passed, the memory cells may degrade, and the patrol read controller (110) can detect the degraded memory cells. For example, the patrol read controller (110) can perform a patrol read on the first user data (UD1) stored in the first memory block (410a) and determine whether the first user data (UD1) is degraded. For example, if the voltage applied to the memory cell where the first user data (UD1) is stored is below a specific voltage, the patrol read controller (110) can determine that the first user data (UD1) is degraded data.

[0064] The reclaim threshold value determiner (120) can determine a threshold value that serves as a criterion for performing a reclaim. For example, the reclaim threshold value determiner (120) can determine a threshold value that serves as a criterion used in the cold data determiner (130). A detailed method for determining the threshold value of the reclaim threshold value determiner (120) will be described later.

[0065] The cold data determiner (130) can determine whether data stored in the non-volatile memory (400) is cold data. Cold data may correspond to data stored in memory cells that are degraded to the extent that reclaiming is required. The cold data determiner (130) can determine user data determined by the patrol read controller (110) as cold data. For example, the cold data determiner (130) can determine user data as cold data by comparing the threshold value determined by the reclaim threshold value determiner (120) with the reclaim count of the user data. A more detailed explanation of this will be provided later.

[0066] The cold data reclaim controller (140) can perform a reclaim on the determined cold data. For example, the cold data reclaim controller (140) can copy the cold data and provide it to another memory block.

[0067] Referring to FIGS. 1, 5, and 6, a host device (20) may provide a write command (WCMD), a logical block address (LBA), and first user data (UD1) to a storage controller (200) (S500). The storage controller (200) may generate a first address (ADDR1) that matches the logical block address (LBA) through a flash translation layer (214). Here, the logical block address (LBA) and the first address (ADDR1) may be matched by a matching table. The storage controller (200) may generate first metadata (MD1) (S501). For example, the storage controller (200) may generate the first metadata (MD1) based on the first user data (UD1). Here, the first metadata (MD1) may include information about the first metadata (MD1) stored in the non-volatile memory (400) and the memory cell array (410) where the first metadata (MD1) is stored, but embodiments of the present invention are not limited thereto. Subsequently, the storage controller (200) may provide a write command (WCMD), first user data (UD1), first metadata (MD1), and a first address (ADDR1) to the non-volatile memory (400) (S502).

[0068] The non-volatile memory (400) can write the first user data (UD1) and the first metadata (MD1) to the first memory block (410a) (S503). For example, the non-volatile memory (400) can program the first user data (UD1) and the first metadata (MD1) to the first memory block (410a) corresponding to the first address (ADDR1). The first user data (UD1) and the first metadata (MD1) can be programmed to different parts of the first memory block (410a). Here, the first metadata (MD1) may be updated by the storage controller (200). Over time, the first user data (UD1) stored in the first memory block (410a) may degrade. That is, retention of the first user data (UD1) may occur.

[0069] After the first user data (UD1) and the first metadata (MD1) are written, the storage controller (200) can perform a patrol read (S510).

[0070] FIG. 7 is a diagram illustrating a patrol lead. FIG. 8 is a flowchart illustrating a reclaim count-up and cold data reclaim according to some embodiments.

[0071] Referring to FIG. 7, the patrol read controller (110) can control the voltage generator (430) of the non-volatile memory (400). For example, the voltage generator (430) can perform a patrol read by applying a monitoring voltage (VM) to the memory cell array (410). As the monitoring voltage (VM) is applied to the first memory block (410a), it can be determined whether the memory cell in which the first user data (UD1) is stored is degraded. That is, depending on the voltage of the memory cell in which the first user data (UD1) is stored, it can be determined whether the first user data (UD1) is degraded.

[0072] Referring to FIG. 8, the reclaim controller (100) can determine whether the first user data (UD1) requires reclaim (S511). That is, the reclaim controller (100) can determine whether the first memory block (410a) in which the first user data (UD1) is stored is degraded. When a monitoring voltage (VM) is applied and the output voltage is different from the reference voltage, the patrol read controller (110) can determine that the corresponding memory cell requires reclaim.

[0073] If the first user data (UD1) requires reclaiming (S511-Y), the reclaim count for the first user data (UD1) may be increased (S512). For example, a reclaim for the first user data (UD1) may have already been performed. If the reclaim count for the first user data (UD1) is n, the reclaim controller (100) may increase the reclaim count for the first user data (UD1) to n+1. That is, the reclaim controller (100) may update the reclaim count for the first user data (UD1).

[0074] If the first user data (UD1) does not require a reclaim (S511-N), the patrol lead may be terminated. That is, if the first user data (UD1) does not require a reclaim, the reclaim count for the first user data (UD1) may not increase.

[0075] After the reclaim count for the first user data (UD1) is increased, the reclaim controller (100) can determine whether the reclaim count for the first user data (UD1) is greater than the reclaim threshold value (S513). Here, the reclaim threshold value can be determined by the reclaim threshold value determiner (120).

[0076] Figure 10 is a flowchart explaining the operation of the threshold value determiner of Figure 5.

[0077] The reclaim threshold determiner (120) can determine a reclaim threshold (S530) prior to performing a patrol read (S510). The reclaim threshold determiner (120) can determine a reclaim threshold based on retention value deviations of multiple memory blocks (410a, 410b), cold data acceptance limits of multiple memory blocks (410a, 410b), and garbage collection overhead status. For example, the reclaim controller (100) can determine a reclaim threshold based on the deletion count of the first memory block (410a) and the deletion count of the second memory block (410b). Accordingly, when the reclaim threshold is small, a large amount of cold data may be determined, and when the reclaim threshold is large, a smaller amount of cold data may be determined.

[0078] If the reclaim count for the first user data (UD1) is greater than the reclaim threshold (S513-Y), the reclaim controller (100) can determine the first user data (UD1) as the first cold data (CD1) and generate the first metadata (MD1') (S520). That is, the reclaim controller (100) can determine whether the first user data (UD1) is cold data by comparing the reclaim count and the reclaim threshold for the first user data (UD1). In other words, the first user data (UD1) can be determined as cold data through the determination of whether the first user data (UD1) requires reclaiming (S511) and the comparison of the reclaim count and the reclaim threshold for the first user data (UD1) (S513). Through this, a storage device (10) with improved performance can be provided.

[0079] Additionally, a first metadata (MD1') including a reclaim count for the first user data (UD1) may be generated. The reclaim controller (100) or the storage controller (200) can improve the performance and lifespan of the storage device (10) by using the first metadata (MD1') including a reclaim count for the first user data (UD1) for wear-leveling, garbage collection, etc. That is, in the field of existing technology, cold data was classified using only a read count for user data, but in an embodiment of the present invention, cold data is classified based on a reclaim count, and by using the reclaim count for operations such as wear-leveling and garbage collection, a storage device (10) with improved performance can be provided.

[0080] For example, the reclaim controller (100) can distribute data to each region of the memory cell array (410) according to the lifespan of each region of the memory cell array (410) based on the first metadata (MD1'). Additionally, in a wear-leveling operation, the reclaim controller (100) can provide data to a memory block with a high erase count.

[0081] The reclaim controller (100) can perform a reclaim on the first cold data (CD1) (S521). For example, the reclaim controller (100) can perform a reclaim on the first user data (UD1) determined to be the first cold data (CD1). Referring again to FIG. 6, the storage controller (200) can provide a write command (WCMD), the first user data (UD1), the first metadata (MD1'), and the second address (ADDR2) to the non-volatile memory (400) (S522). Accordingly, the non-volatile memory (400) can write the first user data (UD1) and the first metadata (MD1') to the second memory block (410b) (S523).

[0082] Figure 9 is a block diagram of a non-volatile memory where reclaiming has been performed.

[0083] Referring to FIG. 9, the first user data (UD1) stored in the first memory block (410a) can be moved to the second memory block (410b) by the cold data reclaim controller (140) of the reclaim controller (100). That is, the first memory block (410a) can delete the first user data (UD1), and the second memory block (410b) can write the first user data (UD1). Additionally, the first metadata (MD1') containing the counted-up reclaim count can be written to the second memory block (410b). Accordingly, the reclaim controller (100) can access the second memory block (410b) to retrieve the first metadata (MD1').

[0084] The storage capacity of the memory cells included in the second memory block (BLK2) may be greater than the storage capacity of the memory cells included in the first memory block (BLK1). That is, if the second memory block (BLK2) is TLC, the first memory block (BLK1) may be SLC. By storing the first user data (UD1) corresponding to the first cold data (CD1) in the second memory block (BLK2) having a higher bit density, a storage device (10) with improved performance may be provided.

[0085] If the reclaim count for the first user data (UD1) is not greater than the reclaim threshold (S513-N), the reclaim controller (100) may perform a reclaim on the first user data (UD1) (S514). In this case, the reclaim controller (100) may not determine the first user data (UD1) as cold data. Therefore, the first metadata (MD1') may not be generated, and the existing first metadata (MD1) may be retained. In this case, the reclaim operation for the first user data (UD1) may differ from the reclaim operation for the first user data (UD1) determined as the first cold data (CD1). For example, the reclaim operation of S514 may correspond to moving the first user data (UD1) to a free block of the memory cell array (410). However, the reclaim operation of S521 may correspond to moving the first user data (UD1) to a second memory block (410b) having a higher bit density.

[0086] FIG. 11 is a diagram showing a change in a matching table according to some embodiments.

[0087] Referring to FIGS. 6 and FIGS. 11, the storage controller (200) can change the matching table (S524). For example, the first matching table (MT1) before the reclaim is performed may contain information about the first address (ADDR1) that matches the logical block address (LBA). The second matching table (MT2) after the reclaim is performed may contain information about the second address (ADDR2) that matches the logical block address (LBA). That is, depending on the reclaim, the storage controller (200) can change the matching table that matches the logical block address (LBA).

[0088] Here, the first address (ADDR1) may correspond to the first memory block (BLK1), and the second address (ADDR2) may correspond to the second memory block (BLK2). That is, as the reclaim count increases and a reclaim for the first cold data (CD1) is performed, a matching table change from the first address (ADDR1) to the second address (ADDR2) may be performed. This may correspond to an interface between the storage controller (200) and the non-volatile memory (400).

[0089] FIG. 12 is a ladder diagram for explaining the read operation of first user data according to some embodiments.

[0090] Referring to FIG. 12, after the matching table is changed (S524), the host device (20) can provide the read command (RCMD) and the logical block address (LBA) to the storage controller (200) (S540). The storage controller (200) can match the logical block address (LBA) to the second address (ADDR2) (S541). That is, the storage controller (200) can derive the second address (ADDR2) according to the second matching table (MT2).

[0091] Next, the storage controller (200) can provide a read command (RCMD) and a second address (ADDR2) to the non-volatile memory (400) (S542). The non-volatile memory (400) can read the first user data (UD1) (S543). For example, the non-volatile memory (400) can read the first user data (UD1) from the second memory block (410b) corresponding to the second address (ADDR2). Here, the first user data (UD1) can be stored in the second memory block (410b) rather than the first memory block (410a).

[0092] The non-volatile memory (400) can provide the first user data (UD1) to the storage controller (200) (S544), and the storage controller (200) can provide the first user data (UD1) to the host device (20) (S545). As the first user data (UD1) corresponding to cold data is stored in the second memory block (410b) with lower performance, the first memory block (410a) with relatively higher performance can be used more efficiently.

[0093] Hereinafter, a storage device (10') according to another embodiment will be described with reference to FIGS. 13 and 14.

[0094] FIGS. 13 and 14 are block diagrams of a storage device comprising a plurality of memory blocks according to some embodiments. For convenience, parts that overlap with those described using FIGS. 1 to 12 are briefly described or omitted.

[0095] Referring to FIG. 13, the storage device (10') may include a reclaim controller (100) and a non-volatile memory (400'). Here, the non-volatile memory (400') may include a plurality of memory blocks. For example, the non-volatile memory (400') may include a first memory block (410a), a second memory block (410b), and a third memory block (410c). Unlike the non-volatile memory (400) described with reference to FIG. 1 to 12, which includes only the first memory block (410a) and the second memory block (410b), the non-volatile memory (400') may further include a third memory block (410c).

[0096] The first memory block (410a) can store first user data (UD1) and first metadata (MD1), and the second memory block (410b) can store second user data (UD2) and second metadata (MD2). Here, the first metadata (MD1) may include a reclaim count for the first user data (UD1), and the second metadata (MD2) may include a reclaim count for the second user data (UD2).

[0097] Referring to FIG. 14, the reclaim controller (100) can perform reclaims on the first user data (UD1) stored in the first memory block (410a) and the second user data (UD2) stored in the second memory block (410b). In this case, the reclaim count for the first user data (UD1) can be increased, and the reclaim count for the second user data (UD2) can be increased. That is, a first metadata (MD1'') containing the increased reclaim count for the first user data (UD1) can be generated, and a second metadata (MD2'') containing the increased reclaim count for the second user data (UD2) can be generated.

[0098] As reclaiming is performed for the first user data (UD1) and the second user data (UD2), the first user data (UD1), the second user data (UD2), the first metadata (MD1''), and the second metadata (MD2'') can be stored in the third memory block (410c). Here, the storage capacity of the memory cells included in the third memory block (410c) may be greater than the storage capacity of the memory cells included in the first memory block (410a) and the second memory block (410b). As the reclaimed first user data (UD1) and the second user data (UD2) are stored in the third memory block (410c), the performance of the storage device (10') may be further improved.

[0099] Hereinafter, a memory system (2) according to another embodiment will be described with reference to FIGS. 15 and FIGS. 16.

[0100] FIG. 15 relates to a memory system comprising a plurality of storage devices according to some embodiments. FIG. 16 is a ladder diagram for explaining the operation of the memory system of FIG. 15. For convenience, parts that overlap with those explained using FIG. 1 to FIG. 12 are briefly explained or omitted.

[0101] Referring to FIG. 15, the memory system (2) may include a host device (20), a first storage device (10a), and a second storage device (10b). The first storage device (10a) and the second storage device (10b) may be separated from each other.

[0102] The first storage device (10a) may include a first storage controller (200a) and a first non-volatile memory (400a). The second storage device (10b) may include a second storage controller (200b) and a second non-volatile memory (400b). Here, the first non-volatile memory (400a) may store first user data (UD1) and first metadata (MD1). The storage capacity of the memory cells included in the first non-volatile memory (400a) may be smaller than the storage capacity of the memory cells included in the second non-volatile memory (400b). That is, the first non-volatile memory (400a) and the second non-volatile memory (400b) may be different from each other.

[0103] Referring to FIGS. 15 and 16, the first non-volatile memory (400a) can write the first user data (UD1) and the first metadata (MD1) (S550). Subsequently, the first storage controller (200a) can perform a patrol read (S551). The first storage controller (200a) can perform a patrol read on the first user data (UD1) stored in the first non-volatile memory (400a). Accordingly, the first storage controller (200a) can determine the first user data (UD1) as the first cold data (CD1) and generate the first metadata (MD1') (S552). At this time, the reclaim count for the first user data (UD1) is increased, and when the reclaim count for the first user data (UD1) is greater than the reclaim threshold value, the storage controller (200a) can perform a reclaim for the first user data (UD1) (S553).

[0104] As a reclaim for the first user data (UD1) is performed, the first storage controller (200a) can provide the first user data (UD1) and the first metadata (MD1') to the host device (20) (S554). The host device (20) can provide the first user data (UD1) and the first metadata (MD1') to the second storage controller (200b) (S555).

[0105] Accordingly, the second storage controller (200b) can provide the first user data (UD1) and the first metadata (MD1') to the second non-volatile memory (400b) (S556). As the reclaim is performed, the second non-volatile memory (400b) can write the first user data (UD1) and the first metadata (MD1') (S557).

[0106] That is, as a reclaim is performed on the first user data (UD1) stored in the first non-volatile memory (400a), the first user data (UD1) can be stored in the second non-volatile memory (400b). Accordingly, the first non-volatile memory (400a) can be used more efficiently.

[0107] FIG. 17 is a drawing of a data center including a storage device according to some embodiments.

[0108] Referring to FIG. 17, the data center (3000) is a facility that collects various data and provides services, and may be referred to as a data storage center. The data center (3000) may be a system for operating search engines and databases, and may be a computing system used by companies such as banks or government agencies. The data center (3000) may include application servers (3100 to 3100n) and storage servers (3200 to 3200m). The number of application servers (3100 to 3100n) and the number of storage servers (3200 to 3200m) may be selected in various ways according to the embodiment, and the number of application servers (3100 to 3100n) and the number of storage servers (3200 to 3200m) may differ from each other.

[0109] The application server (3100) or storage server (3200) may include at least one of a processor (3110, 3210) and memory (3120, 3220). To describe the storage server (3200) as an example, the processor (3210) can control the overall operation of the storage server (3200) and can access the memory (3220) to execute instructions and / or data loaded in the memory (3220). The memory (3220) may be DDR SDRAM (Double Data Rate Synchronous DRAM), HBM (High Bandwidth Memory), HMC (Hybrid Memory Cube), DIMM (Dual In-line Memory Module), Optane DIMM, and / or NVMDIMM (Non-Volatile DIMM). Depending on the embodiment, the number of processors (3210) and the number of memory (3220) included in the storage server (3200) may be selected in various ways. In one embodiment, the processor (3210) and memory (3220) may provide a processor-memory pair. In one embodiment, the number of processors (3210) and memory (3220) may differ. The processor (3210) may include a single-core processor or a multi-core processor. The above description of the storage server (3200) may similarly apply to the application server (3100). Depending on the embodiment, the application server (3100) may not include a storage device (3150). The storage server (3200) may include at least one storage device (3250). The number of storage devices (3250) included in the storage server (3200) may vary depending on the embodiment. The storage device (3250) may include the storage device (10) described with reference to FIGS. 1 through 16.

[0110] Application servers (3100 to 3100n) and storage servers (3200 to 3200m) can communicate with each other through a network (3300). The network (3300) can be implemented using a Fiber Channel (FC) or Ethernet, etc. In this case, the FC is a medium used for relatively high-speed data transmission, and an optical switch providing high performance / high availability can be used. Depending on the access method of the network (3300), the storage servers (3200 to 3200m) can be provided as file storage, block storage, or object storage.

[0111] In one embodiment, the network (3300) may be a storage-dedicated network such as a Storage Area Network (SAN). For example, the SAN may be an FC-SAN that utilizes an FC network and is implemented according to the FC Protocol (FCP). As another example, the SAN may be an IP-SAN that utilizes a TCP / IP network and is implemented according to the iSCSI (SCSI over TCP / IP or Internet SCSI) protocol. In another embodiment, the network (3300) may be a general network such as a TCP / IP network. For example, the network (3300) may be implemented according to protocols such as FC over Ethernet (FCoE), Network Attached Storage (NAS), and NVMe over Fabrics (NVMe-oF).

[0112] Hereinafter, the description will focus on the application server (3100) and the storage server (3200). The description of the application server (3100) may also apply to other application servers (3100n), and the description of the storage server (3200) may also apply to other storage servers (3200m).

[0113] The application server (3100) can store data requested for storage by a user or client in one of the storage servers (3200 to 3200m) via the network (3300). Additionally, the application server (3100) can obtain data requested for reading by a user or client from one of the storage servers (3200 to 3200m) via the network (3300). For example, the application server (3100) can be implemented as a web server or a DBMS (Database Management System), etc.

[0114] The application server (3100) can access memory (3120n) or storage device (3150n) contained in another application server (3100n) via the network (3300), or can access memory (3220-3220m) or storage device (3250-3250m) contained in storage servers (3200-3200m) via the network (3300). Thus, the application server (3100) can perform various operations on data stored in the application servers (3100-3100n) and / or storage servers (3200-3200m). For example, the application server (3100) can execute commands to move or copy data between the application servers (3100-3100n) and / or storage servers (3200-3200m). At this time, data may be moved from the storage devices (3250-3250m) of the storage servers (3200-3200m) through the memories (3220-3220m) of the storage servers (3200-3200m), or directly to the memories (3120-3120n) of the application servers (3100-3100n). Data moving through the network (3300) may be encrypted data for security or privacy.

[0115] To illustrate with an example of a storage server (3200), the interface (3254) can provide a physical connection between the processor (3210) and the controller (3251) and a physical connection between the Network Interconnect (NIC) (3240) and the controller (3251). For example, the interface (3254) can be implemented in a Direct Attached Storage (DAS) manner, which directly connects the storage device (3250) with a dedicated cable. Additionally, for example, the interface (3254) can be implemented in various interface methods such as ATA (Advanced Technology Attachment), SATA (Serial ATA), e-SATA (external SATA), SCSI (Small Computer Small Interface), SAS (Serial Attached SCSI), PCI (Peripheral Component Interconnection), PCIe (PCI express), NVMe (NVM express), IEEE 1394, USB (universal serial bus), SD (secure digital) card, MMC (multi-media card), eMMC (embedded multi-media card), UFS (Universal Flash Storage), eUFS (embedded Universal Flash Storage), and / or CF (compact flash) card interface.

[0116] The storage server (3200) may further include a switch (3230) and a NIC (3240). The switch (3230) may selectively connect the processor (3210) and the storage device (3250) or selectively connect the NIC (3240) and the storage device (3250) under the control of the processor (3210).

[0117] In one embodiment, the NIC (3240) may include a network interface card, a network adapter, etc. The NIC (3240) may be connected to a network (3300) by a wired interface, a wireless interface, a Bluetooth interface, an optical interface, etc. The NIC (3240) may include internal memory, a Digital Signal Processor (DSP), a host bus interface, etc., and may be connected to a processor (3210) and / or a switch (3230), etc., through the host bus interface. The host bus interface may be implemented as one of the examples of the interface (3254) described above. In one embodiment, the NIC (3240) may be integrated with at least one of a processor (3210), a switch (3230), and a storage device (3250).

[0118] In storage servers (3200-3200m) or application servers (3100-3100n), a processor may program or read data by sending a command to a storage device (3150-3150n, 3250-3250m) or memory (3120-3120n, 3220-3220m). The data may be data that has been error-corrected through an Error Correction Code (ECC) engine. The data may be data that has undergone Data Bus Inversion (DBI) or Data Masking (DM) processing and may include Cyclic Redundancy Code (CRC) information. The data may be data that has been encrypted for security or privacy.

[0119] The storage device (3150-3150n, 3250-3250m) can transmit a control signal and a command / address signal to the NAND flash memory device (3252-3252m) in response to a read command received from the processor. Accordingly, when reading data from the NAND flash memory device (3252-3252m), the RE (Read Enable) signal is input as a data output control signal and can serve to output data to the DQ bus. A DQS (Data Strobe) can be generated using the RE signal. The command and address signals can be latched in the page buffer according to the rising edge or falling edge of the WE (Write Enable) signal.

[0120] The controller (3251) can control the overall operation of the storage device (3250). In one embodiment, the controller (3251) may include a Static Random Access Memory (SRAM). The controller (3251) can write data to the NAND flash (3252) in response to a write command, or read data from the NAND flash (3252) in response to a read command. For example, the write command and / or read command may be provided by a processor (3210) in the storage server (3200), a processor (3210m) in another storage server (3200m), or a processor (3110, 3110n) in the application server (3100, 3100n). The DRAM (3253) may temporarily store (buffer) data to be written to the NAND flash (3252) or data read from the NAND flash (3252). Additionally, the DRAM (3253) can store metadata. Here, the metadata is data generated by the controller (3251) to manage user data or NAND flash (3252). The storage device (3250) may include a Secure Element (SE) for security or privacy.

[0121] Although embodiments of the present invention have been described above with reference to the attached drawings, the present invention is not limited to the above embodiments and can be manufactured in various different forms, and those skilled in the art will understand that the present invention can be implemented in other specific forms without changing the technical concept or essential features of the present invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. Explanation of the symbols

[0122] 1: Memory System 10: Storage Device 20: Host device 100: Reclaim controller 110: Patrol Lead Controller 120: Reclaim Threshold Determiner 130: Cold Data Determiner 140: Cold Data Reclaim Controller 200: Storage controller 400: Non-volatile memory 410a: 1st memory block 410b: 2nd memory block

Claims

Claim 1 A storage device comprising: a non-volatile memory including a first memory block for storing first user data; and a controller for performing a patrol read on the first user data, wherein the controller increases a reclaim count for the first user data when a reclaim for the first user data is required, and when the reclaim count for the first user data is greater than a reclaim threshold, determines the first user data as cold data and performs a first reclaim on the first user data determined as cold data, wherein the non-volatile memory stores the first user data in a second memory block different from the first memory block in response to the first reclaim, and the controller determines the reclaim threshold based on the delete count of the first memory block and the delete count of the second memory block. Claim 2 A storage device according to claim 1, wherein the first memory block comprises SLC (single level cell) and the second memory block comprises at least one of MLC (multi level cell), TLC (triple level cell), and QLC (quad level cell). Claim 3 A storage device according to claim 1, wherein the storage capacity of the first memory cell included in the first memory block is smaller than the storage capacity of the second memory cell included in the second memory block. Claim 4 A storage device according to claim 1, wherein the non-volatile memory includes a third memory block storing second user data, and the controller performs a patrol read on the second user data, increases a reclaim count for the second user data when a reclaim for the second user data is required, and performs a second reclaim on the second user data without determining the second user data as cold data when the reclaim count for the second user data is not greater than the reclaim threshold value. Claim 5 A storage device according to claim 4, wherein the non-volatile memory stores the second user data in a fourth memory block different from the third memory block in response to the second reclaim, and the storage capacity of the third memory cell included in the third memory block is not smaller than the storage capacity of the fourth memory cell included in the fourth memory block. Claim 6 delete Claim 7 A storage device according to claim 1, wherein the non-volatile memory includes a second memory block storing second user data different from the first user data and a third memory block different from the first and second memory blocks, and the controller performs a patrol read on the second user data, and if a reclaim for the second user data is required, increases a reclaim count for the second user data, and if the reclaim count for the second user data is greater than the reclaim threshold value, determines the second user data as cold data, performs a first reclaim on the first user data determined as cold data, and performs a second reclaim on the second user data determined as cold data. Claim 8 Non-volatile memory; A storage device comprising: a controller that receives first user data, a logical block address, and a write command from an external device, generates a first matching table for a first address that matches the logical block address, and provides the first address and the first user data to the non-volatile memory; wherein the non-volatile memory stores the first user data in a first memory block based on the first address; wherein the controller performs a patrol read on the first user data, increases a reclaim count for the first user data if reclaiming of the first user data is required, determines the first user data as cold data if the reclaim count is greater than a reclaim threshold, performs a reclaim on the first user data determined as cold data, generates a second matching table for a second address that matches the logical block address, wherein the second address is different from the first address; and wherein the controller determines the reclaim threshold based on the erase count of the first memory block and the erase count of the second memory block where the first user data is stored based on the second address. Device. Claim 9 In claim 8, the controller performs a reclaim for the first user data based on the reclaim count, and the first memory block in which the first user data is stored based on the first address is a storage device different from the second memory block in which the first user data is stored based on the second address. Claim 10 A storage device comprising: a first non-volatile memory for storing first user data; a second non-volatile memory different from the first non-volatile memory; and a controller for performing a patrol read on the first user data, wherein the controller increases a reclaim count for the first user data when reclaiming the first user data is required, and when the reclaim count for the first user data is greater than a reclaim threshold, the first user data is provided to the second non-volatile memory, the second non-volatile memory stores the first user data, and the controller determines the reclaim threshold based on the erase count of the first non-volatile memory and the erase count of the second non-volatile memory.

Citation Information

Patent Citations

  • Control device of storage system and storage space recovery method of storage system

    CN107436729A

  • Memory processing method and device, electronic equipment and computer readable storage medium

    CN109992523A

  • Memory recovery processing method and device, electronic equipment and storage medium

    CN110764906A

  • Test case execution method and device, execution equipment and readable storage medium

    CN112000586A

  • Memory management method and electronic equipment

    CN112817736A