Memory system and controller of memory system

The memory system optimizes recovery operations by managing logical regions with a write state table, reducing recovery time and maintaining performance through selective recovery, addressing inefficiencies in existing systems.

KR103047690B1Active Publication Date: 2026-11-11SK HYNIX INC
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
KR1020200169157
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2020-12-07
Publication Date
2026-11-11
Estimated Expiration
2040-12-07

AI Technical Summary

Technical Problem

Existing memory systems face inefficiencies in performing recovery operations following abnormal terminations, leading to prolonged recovery times and performance degradation.

Method used

A memory system with a controller that manages logical and physical regions using a write state table, allowing selective recovery operations based on the state of logical regions, thereby omitting unnecessary operations and reducing recovery time.

Benefits of technology

The solution enables efficient recovery operations by minimizing unnecessary recovery tasks, shortening recovery time, and maintaining system performance even after abnormal shutdowns.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112020131960657-PAT00001_ABST
    Figure 112020131960657-PAT00001_ABST
Patent Text Reader

Abstract

A memory system comprises a storage medium including a plurality of physical regions; and a controller configured to map logical regions configured by a host device to the physical regions and to perform a write operation on the physical region mapped to the target logical region in response to a write request for the target logical region, wherein the controller is configured to update a write state corresponding to the target logical region in a write state table in response to the write request.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a memory system, and more specifically, to a memory system including a non-volatile memory device. Background Technology

[0002] The memory system may be configured to store data provided by the host device in response to a write request from the host device. Additionally, the memory system may be configured to provide the stored data to the host device in response to a read request from the host device. The host device is an electronic device capable of processing data and may include a computer, a digital camera, or a mobile phone. The memory system may operate by being embedded in the host device or by being manufactured in a detachable form and connected to the host device. The problem to be solved

[0003] An embodiment of the present invention provides a memory system and a controller of the memory system that efficiently perform recovery operations for abnormal termination. means of solving the problem

[0004] A memory system according to an embodiment of the present invention comprises: a storage medium including a plurality of physical regions; and a controller configured to map logical regions configured by a host device to the physical regions and to perform a write operation on the physical region mapped to the target logical region in response to a write request for the target logical region, wherein the controller may include a controller configured to update a write state corresponding to the target logical region in a write state table in response to the write request.

[0005] A memory system according to an embodiment of the present invention may include a storage medium comprising a plurality of physical regions; and a controller configured to respond to a write request for a target physical region among the physical regions, change the first state to a second state when the write state corresponding to the target physical region in a write state table is the first state, and maintain the second state when the write state is the second state.

[0006] A controller of a memory system according to an embodiment of the present invention may include: a write unit configured to map logical regions configured by a host device to physical regions and to perform a write operation on a physical region mapped to a target logical region in response to a write request for a target logical region; and a recovery unit configured to omit a recovery operation for a physical region mapped to a logical region in a first state of write state and to perform a recovery operation for a physical region mapped to a logical region in a second state of write state when it is determined that there was an abnormal termination immediately prior to the boot operation. Effects of the invention

[0007] A memory system and a controller of a memory system according to an embodiment of the present invention can efficiently perform a recovery operation. Brief explanation of the drawing

[0008] FIG. 1 is a block diagram illustrating a data processing system according to an embodiment of the present invention. FIG. 2 is a diagram for explaining logical regions and physical regions according to an embodiment of the present invention, FIG. 3 is a diagram illustrating a method for updating a light state in a light state table stored in a host memory buffer according to an embodiment of the present invention. FIG. 4 is a drawing illustrating a method of performing a recovery operation by referring to the light status of a light status table stored in a host memory buffer according to an embodiment of the invention. FIG. 5 is a flowchart illustrating the operation method of the light unit of FIG. 1 according to an embodiment of the present invention. FIG. 6 is a flowchart illustrating the operation method of the recovery unit of FIG. 1 according to an embodiment of the present invention. FIG. 7 is a flowchart illustrating the operation method of the recovery unit of FIG. 1 according to an embodiment of the present invention. FIG. 8 is a drawing exemplarily illustrating a data processing system including a solid-state drive (SSD) according to an embodiment of the present invention, FIG. 9 is a drawing exemplarily illustrating a data processing system including a memory system according to an embodiment of the present invention, FIG. 10 is a drawing exemplarily illustrating a data processing system including a memory system according to an embodiment of the present invention, FIG. 11 is a drawing exemplarily illustrating a network system including a memory system according to an embodiment of the present invention, FIG. 12 is a block diagram exemplarily illustrating a non-volatile memory device included in a memory system according to an embodiment of the present invention. Specific details for implementing the invention

[0009] The advantages and features of the present invention and the methods for achieving them will be explained in detail through the embodiments described below in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments described herein and may be embodied in other forms. These embodiments are provided merely to explain in detail enough to enable those skilled in the art to easily implement the technical concept of the present invention.

[0010] In the drawings, embodiments of the invention are not limited to the specific forms depicted and are exaggerated for clarity. Specific terms have been used herein, but are used for the purpose of describing the invention and are not intended to limit the meaning or the scope of the invention as defined in the claims.

[0011] In this specification, the expression "and / or" is used to mean including at least one of the components listed before and after it. Additionally, the expression "connected / combined" is used to mean being directly connected to another component or being indirectly connected through another component. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. Furthermore, components, steps, actions, and elements referred to as "include" or "comprising" in this specification signify the presence or addition of one or more other components, steps, actions, and elements.

[0012] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0013] FIG. 1 is a block diagram illustrating a data processing system (10) according to an embodiment of the present invention.

[0014] Referring to FIG. 1, the data processing system (10) is an electronic system capable of processing data and may include a data center, an internet data center, a cloud data center, a personal computer, a laptop computer, a smartphone, a tablet computer, a digital camera, a game console, a navigation system, a virtual reality device, and a wearable device.

[0015] The data processing system (10) may include a host device (100) and a memory system (200).

[0016] The host device (100) can access the memory system (200) using logical addresses while processing data. The host device (100) can store data in the memory system (200) by assigning logical addresses to the data.

[0017] The host device (100) may include a logical area management unit (110), a processor (120), and a host memory buffer (130).

[0018] The logical region management unit (110) can configure multiple logical regions based on logical addresses. Each logical region can be composed of consecutive logical addresses. The logical region management unit (110) can allocate a logical region that is not in use (i.e., a logical region in which the logical addresses constituting the logical region are not assigned to data) to the processor (120) in response to a request from the processor (120).

[0019] The processor (120) may request the logical area management unit (110) to allocate a logical area to store data in the memory system (200). After the logical area is allocated to the processor (120), the processor (120) may store data in the memory system (200) using the logical addresses of the logical area. Specifically, the processor (120) may store data in the memory system (200) by allocating the logical addresses of the logical area to data and then transmitting a write request containing the logical address and data to the memory system (200). In the following, when the processor (120) transmits a write request for a logical address and data to the memory system (200), the logical area containing the corresponding logical address will be referred to as the target logical area of ​​the write request.

[0020] Meanwhile, the processor (120) can use consecutive logical addresses constituting a logical region in order from the first logical address to the last logical address. Accordingly, a write request generated by the processor (120) may be a sequential write request for consecutive logical addresses.

[0021] The processor (120) may be allocated one logical region at a time. According to an embodiment, the processor (120) may be allocated two or more logical regions simultaneously. Additionally, although not illustrated, the host device (100) may include multiple processors, and each of the processors may be allocated a logical region. In this case, the write requests generated by one or more processors may be for different target logical regions, and each of the write requests may be a sequential write request.

[0022] The processor (120) may include a central processing unit, a graphics processing unit, a microprocessor (120), an application processor (120), an acceleration processing unit, or an operating system.

[0023] The host memory buffer (130) may store a write state table (135). The write state table (135) may be managed and referenced by the write section (211) and the recovery section (212) of the memory system (200) as described below. The write state table (135) may include write states corresponding to one or more logical regions.

[0024] The host memory buffer (130) may be supported according to, for example, the NVMe (NVM Express) standard.

[0025] The memory system (200) may be configured to store data provided from the host device (100) in response to a write request from the host device (100). Additionally, the memory system (200) may be configured to provide the stored data to the host device (100) in response to a read request from the host device (100).

[0026] The memory system (200) may include a PCMCIA (Personal Computer Memory Card International Association) card, a CF (Compact Flash) card, a smart media card, a memory stick, various multimedia cards (MMC, eMMC, RS-MMC, MMC-micro), an SD (Secure Digital) card (SD, Mini-SD, Micro-SD), a UFS (Universal Flash Storage), or an SSD (Solid State Drive).

[0027] The memory system (200) may include a controller (210) and a storage medium (220).

[0028] The controller (210) can control the general operations of the memory system (200). The controller (210) can control the storage medium (220) to perform foreground operations according to instructions from the host device (100). Foreground operations may include operations to write data to the storage medium (220) and read data from the storage medium (220) according to instructions from the host device (100), namely, write requests and read requests.

[0029] Additionally, the controller (210) can control the storage medium (220) to perform necessary background operations internally independently of the host device (100). The background operations may include at least one of wear leveling operations, garbage collection operations, erase operations, read reclaim operations, and refresh operations for the storage medium (220). The background operations may include operations to write data to the storage medium (220) and to read data from the storage medium (220), similar to foreground operations.

[0030] The controller (210) may include a light section (211) and a recovery section (212).

[0031] The light unit (211) can map a logical region allocated to the processor (120) to one or more physical regions of the storage medium (220). The light unit (211) can manage mapping information between the logical region and the physical region. The light unit (211) can store data in the physical region to which the target logical region of the light request is mapped in response to a light request transmitted from the processor (120). According to an embodiment, the light unit can map the logical region to one or more physical regions when it first receives a light request for the corresponding logical region for each logical region.

[0032] The write unit (211) can manage a write state table (135) in the host memory buffer (130). The write unit (211) can update the write state of the target logical region (i.e., the write state corresponding to the logical region identifier of the target logical region) in the write state table (135) stored in the host memory buffer (130) in response to a write request for the target logical region. Specifically, when the write state of the target logical region is a first state (e.g., disabled state), the write unit (211) can change the disabled state to a second state (e.g., enabled state). When the write state of the corresponding logical region is an enabled state, the write unit (211) can maintain the enabled state.

[0033] According to an embodiment, the light unit (211) can update the light status of the target logical area before, during, or after performing a light operation after receiving a light request for the target logical area.

[0034] According to an embodiment, the light unit (211) may manage the light status only for a logical area mapped to an open physical area among the physical areas included in the storage medium (220). An open physical area may be a physical area where a light operation is performed in response to a light request. According to an embodiment, when an open physical area becomes full of data, the light unit (211) may remove the light status of the logical area mapped to the physical area from the light status table (135) or change it to a de-table state.

[0035] According to an embodiment, the light unit (211) may manage the light status only for one or more logical regions that have been the target logical regions of a light request. According to an embodiment, the light unit (211) may manage the light status only for logical regions in use (i.e., logical regions where logical addresses constituting the logical regions are assigned to data). According to an embodiment, the light unit (211) may manage the light status for all logical regions configured by the logical region management unit (110). The logical region management unit (110) may provide the light unit (211) with information about the logical regions.

[0036] According to an embodiment, when the light unit (211) first receives a light request for a logical area when the light state table (135) does not contain a light state for a logical area, the light state for that logical area can be added to the light state table (135) as an enabled state.

[0037] According to an embodiment, when the host device (100) transmits a write request, it may transmit not only a logical address and data but also a logical area identifier (ID) of the target logical area. Accordingly, the write unit (211) can determine the target logical area of ​​the write request by verifying the logical area identifier (ID) included in the write request. According to an embodiment, even if the write request does not include a logical area identifier (ID), the write unit (211) can determine the target logical area based on the logical address included in the write request and can also calculate the logical area identifier (ID).

[0038] The recovery unit (212) can perform a recovery operation when it is determined that there was an abnormal termination immediately prior to the boot operation of the controller (210). At this time, the recovery unit (212) can perform a recovery operation only for selected physical regions in the storage medium (220) by referring to the write state table (135) stored in the host memory buffer (130).

[0039] Specifically, the recovery unit (212) may perform a recovery operation on a physical area mapped to a logical area where the light status is enabled by referring to the light status table (135). The recovery unit (212) may not perform a recovery operation on a physical area mapped to a logical area where the light status is disabled by referring to the light status table (135) (or may skip / omit the recovery operation).

[0040] After performing a recovery operation on a physical area, the recovery unit (212) can change the light status of the logical area mapped to the physical area to a disabled state. Therefore, even if an abnormal termination is repeated, the recovery unit (212) can confirm that the corresponding light status in the light status table (135) for the physical area where the recovery operation was previously performed is disabled. Therefore, unnecessary recovery operations for the physical area where the recovery operation was previously performed can be avoided.

[0041] In summary, according to the present invention, by managing the light status table (135), unnecessary recovery operations for open physical regions that do not require recovery operations can be omitted. That is, since recovery operations are performed only on selected open physical regions among all open physical regions by referring to the light status table (135), the recovery time can be shortened. Furthermore, since the controller (210) can simply identify the open physical regions among all open physical regions for which recovery operations need to be performed by referring to the light status table (135), the analysis time for the recovery operation can be effectively shortened.

[0042] Additionally, even if an abnormal shutdown occurs in the memory system (200), if the power of the host device (100) is maintained, the write state table (135) stored in the host memory buffer (130) can be safely maintained. Therefore, a recovery operation based on the write state table (135) can be properly performed.

[0043] Also, since the light status table (135) is managed in a host memory buffer (130) with a fast operating speed, even if updates to the light status table (135) are performed frequently, it may not affect the operating performance of the memory system (200).

[0044] In addition, the controller (210) manages the light state table (135) in units of logical regions, so the light state can be easily managed regardless of how the physical region is mapped to the logical region in units or in any way.

[0045] According to an embodiment, considering the case where both the memory system (200) and the host device (100) are abnormally terminated, the write unit (211) may periodically store validation information (e.g., parity data) regarding the write state table (135) in the host memory buffer (130) while managing the write state table (135). According to an embodiment, the write unit (211) may generate new validation information for the write state table (135) and store it in the host memory buffer (130) whenever the write state table (135) is updated. The recovery unit (212) may prioritize determining the validity of the write state table (135) based on the validation information when it is determined that there was an abnormal termination of the memory system (200) during the boot operation. When the write state table (135) is determined to be valid, the recovery unit (212) may check the write state table (135) and perform the above-described recovery operation only on selected open physical regions. The recovery unit (212) may ignore the light state table (135) and perform a recovery operation when it is determined that the light state table (135) is invalid. According to an embodiment, the recovery unit (212) may perform a recovery operation for all one or more open physical regions when it is determined that the light state table (135) is invalid.

[0046] According to an embodiment, the write unit (211) may manage not only the write status for a logical area in the write status table (135), but also the write status for a system area of ​​the storage medium (220). The system area may be a physical area that stores data necessary for the operation of the memory system (200) and data generated during operation, rather than data transmitted from the host device (100). Therefore, the system area may not be mapped to a separate logical area. Accordingly, the write unit (211) may manage the write status for the system area in the write status table (135) in order to efficiently manage the need for a recovery operation for the system area. The method of managing the write status for the system area as an enable state or a disable state may be the same as the method of managing the write status for the logical area. And the recovery unit (212) may selectively perform a recovery operation for the system area by referring to the write status for the system area as described above.

[0047] The storage medium (220) can store data transmitted from the controller (210) under the control of the controller (210), and read the stored data to transmit it to the controller (210).

[0048] The controller (210) may configure multiple physical regions in the storage medium (220). Each physical region may be a unit in which an erase operation is performed in the storage medium (220). In other words, all data stored in each physical region may be erased together. For example, each physical region may be a memory block contained in a single non-volatile memory device. As another example, each physical region may be a set of memory blocks contained in a plurality of non-volatile memory devices and capable of parallel access. Each physical region may be, for example, a super block.

[0049] Each physical region may include a plurality of memory units corresponding to consecutive physical addresses. Each memory unit may be a unit in which a write operation or a read operation is performed on the storage medium (220). The write unit (211) may store data in the memory units in the order of physical addresses.

[0050] When data is stored in a memory unit in the physical area, the write unit (211) can additionally manage mapping information between the logical address of the data and the physical address of the memory unit.

[0051] The storage medium (220) may include one or more non-volatile memory devices. The non-volatile memory devices may include flash memory devices such as NAND flash or NOR flash, Ferroelectric Random Access Memory (FeRAM), Phase-Change Random Access Memory (PCRAM), Magnetic Random Access Memory (MRAM), or Resistive Random Access Memory (ReRAM).

[0052] FIG. 2 is a diagram illustrating logic regions (LR) and physical regions (PR) according to an embodiment of the present invention.

[0053] Referring to FIG. 2, the host device (100) can divide the logical addresses (LA) from "0" to "i" to form one or more logical regions (LR). Each of the logical regions (LR) can be composed of consecutive logical addresses.

[0054] Each of the logical regions (LR) may be mapped to one or more physical regions (PR) of the storage medium (220). For example, as illustrated, the logical regions (LR0, LR1, LR2, LR3) may be mapped to the physical regions (PR0, PR1, PR2, PR3), respectively. Meanwhile, FIG. 2 illustrates an embodiment in which one logical region (LR) is mapped to one physical region (PR), but one logical region (LR) may be mapped to two or more physical regions (PR). Also, FIG. 2 illustrates an embodiment in which the logical regions (LR) and physical regions (PR) are mapped sequentially in address order, but the logical regions (LR) and physical regions (PR) may be mapped regardless of address order.

[0055] According to an embodiment, the write unit (211) may manage mapping information between logical regions (LR) and physical regions (PR). For example, the mapping information may include a logical region identifier of a logical region (LR) and a physical region address of a physical region (PR) mapped to the corresponding logical region (LR). Thus, the write unit (211) may process write requests and read requests transmitted from the host device (100) by verifying the mapping relationship between the logical region (LR) and the physical region (PR) through the mapping information. According to an embodiment, the write unit (211) may verify the mapping relationship between the logical regions (LR) and the physical regions (PR) according to a predetermined rule without managing or using separate mapping information.

[0056] Logical regions (LR) and physical regions (PR) can be mapped to the same size. That is, the size of each logical region (LR) can be the same as the size of one or more physical regions (PR) mapped to each logical region. In other words, the data size corresponding to the logical addresses constituting each logical region (LR) can be the same as the data size that can be stored in one or more physical regions (PR) mapped to each logical region (LR).

[0057] As described above, the write unit (211) can process a write request for a target logical area (LR) for a physical area (PR) mapped to the target logical area (LR). For example, when the write request is for one or more logical addresses between "0" and "k", the target logical area may be a logical area (LR0). Then, the write unit (211) can store data in the physical area (PR0) mapped to the target logical area (LR0) in response to the write request.

[0058] Meanwhile, in FIG. 2, the physical region (PR3) is filled with data, and each of the physical regions (PR0, PR1, PR2) may be an "open" physical region that is being filled with data. Similarly, the logical region (LR3) mapped to the physical region (PR3) may have all logical addresses assigned to data, and each of the logical regions (LR0, LR1, LR2) mapped to the physical regions (PR0, PR1, PR2) may have logical addresses being assigned to data. FIG. 3 and FIG. 4 will be described below assuming the mapping state of FIG. 2.

[0059] FIG. 3 is a diagram illustrating a method for updating a light state (WS) in a light state table (135) stored in a host memory buffer (130) according to an embodiment of the present invention. FIG. 3 assumes a case where the memory system (200) is powered on after normal shutdown.

[0060] Referring to FIG. 3, at time (T0), the memory system (200) can be powered on. The recovery unit (212) can determine that there was no abnormal shutdown of the memory system (200), such as a sudden power off, immediately before the boot operation.

[0061] At time (T1), unless the host device (100) was also powered off immediately prior, the host memory buffer (130) may maintain a light state table (135) stored by the light unit (211). The light state table (135) may include a logical area identifier (ID) and a light state (WS) corresponding to each logical area identifier (ID). The light state (WS) corresponding to each logical area identifier (ID) may be a light state (WS) for the logical area (LR) pointed to by each logical area identifier (ID). In FIG. 3, as a light state (WS), "0" means a disabled state and "1" means an enabled state, but depending on the embodiment, any other values ​​may be used as a disabled state and an enabled state.

[0062] At the continuing point (T1), the recovery unit (212) may change all write states (WS) of the write state table (135) to a disabled state. The disabled state may be an initial state. The disabled state may mean that no write requests for the corresponding logical area have been processed since power-on. In other words, the disabled state may mean that no write operations have been performed on the physical area mapped to the corresponding logical area since power-on.

[0063] At time (T2), the writer (211) may receive a write request for the target logical region (LR1). Accordingly, the writer (211) may store data in the physical region (PR1) mapped to the target logical region (LR1) in response to the write request.

[0064] At time (T3), the writer (211) can change the write state (WS) of the target logical area (LR1) in the write state table (135) from disabled to enabled. If, while the write state (WS) of the logical area (LR1) is in the enabled state, an additional write request for the logical area (LR1) is subsequently received, the writer (211) can maintain the write state (WS) of the logical area (LR1) in the enabled state. That is, the enabled state may mean that a write request for the logical area (LR1) has been processed since power-on. In other words, the enabled state may mean that a write operation has been performed on the physical area (PR1) mapped to the logical area (LR1) since power-on.

[0065] According to an embodiment, the light state table (135) may include a physical area address instead of a logical area identifier (ID). The light state (WS) corresponding to the physical area address may be the light state (WS) for the corresponding physical area. The meaning of a disabled state or an enabled state as a light state (WS) for the corresponding physical area may be the same as described above. That is, a disabled state may mean that no light operation has been performed on the corresponding physical area since power-on. An enabled state may mean that a light operation has been performed on the corresponding physical area since power-on.

[0066] FIG. 4 is a diagram illustrating a method of performing a recovery operation by referring to the write status (WS) of a write status table (135) stored in a host memory buffer (130) according to an embodiment of the invention. FIG. 4 assumes a case where the memory system (200) is powered on after an abnormal shutdown.

[0067] Referring to FIG. 4, at time (T10), the memory system (200) can be powered on. The recovery unit (212) can determine that there was an abnormal shutdown of the memory system (200), such as a sudden power off, immediately before the boot operation.

[0068] At time point (T11), if the host device (100) was not powered off immediately before, the host memory buffer (130) can maintain the write state table (135) stored by the write unit (211). The recovery unit (212) can check the write state table (135) stored in the host memory buffer (130). The recovery unit (212) can check that in the write state table (135), the write state (WS) of the logical regions (LR1, LR2) is in an enabled state and the write state (WS) of the logical region (LR0) is in a disabled state. An enabled state may mean that if a sudden power off had occurred immediately before, a recovery operation must be performed on the physical regions (PR1, PR2) mapped to the logical regions (LR1, LR2). The disabled state may mean that there is no need to perform a recovery operation on the physical area (PR0) mapped to the logical area (LR0), even if a sudden power off occurred immediately before.

[0069] At time point (T12), the recovery unit (212) may perform a recovery operation on physical regions (PR1, PR2) mapped to logical regions (LR1, LR2). The recovery operation may include at least one of a dummy write operation for an unstable memory unit, a copy operation for unstable data, a map data recovery operation, and various metadata recovery operations in each of the physical regions (PR1, PR2). Meanwhile, the recovery unit (212) may not perform a recovery operation on the physical region (PR0) mapped to the logical region (LR0).

[0070] At time point (T13), the recovery unit (212) can change the write state (WS) of the logical regions (LR1, LR2) in the host memory buffer (130) from an enabled state to a disabled state. That is, the disabled state may mean that the recovery operation for the physical regions (PR1, PR2) mapped to the logical regions (LR1, LR2) has been completed, even if a sudden power off occurred immediately before.

[0071] Meanwhile, the boot operation can be performed not only after the memory system (200) is powered on, but also when the host device (100) instructs the memory system (200) to perform a hardware reset. Accordingly, although the time points (T0, T10) of FIGS. 3 and 4 represent the case where the memory system (200) is powered on, the recovery unit (212) can operate in the same manner as described with reference to FIGS. 3 and 4 even when the host device (100) instructs the memory system (200) to perform a hardware reset.

[0072] FIG. 5 is a flowchart illustrating the operation method of the light unit (211) of FIG. 1 according to an embodiment of the present invention.

[0073] Referring to FIG. 5, in step (S110), the light unit (211) can receive a light request for a target logic area.

[0074] In step (S120), the light unit (211) can check the light status of the target logical area in the light status table (135) stored in the host memory buffer (130).

[0075] In step (S130), the light unit (211) can determine whether the light state of the target logic area is an enabled state. When it is determined that the light state is an enabled state, the procedure can proceed to step (S140). When it is determined that the light state is a disabled state, the procedure can proceed to step (S150).

[0076] In step (S140), the light unit (211) can maintain an enabled state as the light state of the target logic area.

[0077] In step (S150), the light unit (211) can change the light state of the target logic area from disabled state to enabled state.

[0078] FIG. 6 is a flowchart illustrating the operation method of the recovery unit (212) of FIG. 1 according to an embodiment of the present invention.

[0079] Referring to FIG. 6, in step (S210), the controller (210) can initiate a boot operation.

[0080] In step (S220), the recovery unit (212) can determine whether there was an abnormal shutdown of the memory system (200) immediately prior. If it is determined that there was no abnormal shutdown, the procedure can proceed to step (S260). If it is determined that there was an abnormal shutdown, the procedure can proceed to step (S230).

[0081] In step (S230), the recovery unit (212) can identify a logical region in which the light state is enabled in the light state table (135) stored in the host memory buffer (130).

[0082] In step (S240), the recovery unit (212) can perform a recovery operation on a physical area mapped to an identified logical area. At this time, the recovery unit (212) may not perform a recovery operation on a physical area mapped to a logical area where the light status in the light status table (135) is disabled. Therefore, the time required to perform the recovery operation can be significantly reduced.

[0083] In step (S250), the recovery unit (212) can change the light state of the identified logical area from an enabled state to a disabled state.

[0084] In step (S260), the recovery unit (212) can change all light states in the light state table (135) stored in the host memory buffer (130) to a disabled state.

[0085] FIG. 7 is a flowchart illustrating the operation method of the recovery unit (212) of FIG. 1 according to an embodiment of the present invention.

[0086] Referring to FIG. 7, in step (S310), the controller (210) can initiate a boot operation.

[0087] In step (S320), the recovery unit (212) can determine whether there was an abnormal shutdown of the memory system (200) immediately prior. If it is determined that there was no abnormal shutdown, the procedure can proceed to step (S370). If it is determined that there was an abnormal shutdown, the procedure can proceed to step (S330).

[0088] In step (S330), the recovery unit (212) can determine whether the light status table (135) is valid based on the validation information stored in the host memory buffer (130). When it is determined that the light status table (135) is valid, the procedure can proceed to step (S340). When it is determined that the light status table (135) is invalid, the procedure can proceed to step (S380).

[0089] In step (S340), the recovery unit (212) can identify a logical region in which the light state is enabled in the light state table (135) stored in the host memory buffer (130).

[0090] In step (S350), the recovery unit (212) can perform a recovery operation on a physical area mapped to an identified logical area. At this time, the recovery unit (212) may not perform a recovery operation on a physical area mapped to a logical area where the light status in the light status table (135) is disabled. Therefore, the time required to perform the recovery operation can be significantly reduced.

[0091] In step (S360), the recovery unit (212) can change the light state of the identified logical area from an enabled state to a disabled state.

[0092] In step (S370), the recovery unit (212) can change all light states in the light state table (135) stored in the host memory buffer (130) to a disabled state.

[0093] In step (S380), the recovery unit (212) can perform a recovery operation on all open physical regions.

[0094] FIG. 8 is a diagram illustrating an exemplary data processing system including a solid state drive (SSD) according to an embodiment of the present invention. Referring to FIG. 8, the data processing system (1000) may include a host device (1100) and a solid state drive (1200) (hereinafter referred to as SSD).

[0095] The SSD (1200) may include a controller (1210), a buffer memory device (1220), non-volatile memory devices (1231 to 123n), a power supply (1240), a signal connector (1250), and a power connector (1260).

[0096] The controller (1210) can control the general operation of the SSD (1200). The controller (1210) may include a host interface unit (1211), a control unit (1212), a random access memory (1213), an error correction code (ECC) unit (1214), and a memory interface unit (1215).

[0097] The host interface unit (1211) can exchange signals (SGL) with the host device (1100) through the signal connector (1250). Here, the signal (SGL) may include commands, addresses, data, etc. The host interface unit (1211) can interface the host device (1100) with the SSD (1200) according to the protocol of the host device (1100). For example, the host interface unit (1211) can communicate with the host device (1100) through any one of the standard interface protocols such as secure digital, USB (universal serial bus), MMC (multi-media card), eMMC (embedded MMC), PCMCIA (personal computer memory card international association), PATA (parallel advanced technology attachment), SATA (serial advanced technology attachment), SCSI (small computer system interface), SAS (serial attached SCSI), PCI (peripheral component interconnection), PCI-E (PCI Expresss), and UFS (universal flash storage).

[0098] The control unit (1212) can analyze and process a signal (SGL) input from the host device (1100). The control unit (1212) can control the operation of background function blocks according to firmware or software for driving the SSD (1200). Random access memory (1213) can be used as an operating memory for driving such firmware or software.

[0099] The control unit (1212) may include the light unit (211) and the recovery unit (212) of FIG. 1.

[0100] The error correction code (ECC) unit (1214) can generate parity data for data to be transmitted to non-volatile memory devices (1231–123n). The generated parity data can be stored in the non-volatile memory devices (1231–123n) along with the data. The error correction code (ECC) unit (1214) can detect errors in the data read from the non-volatile memory devices (1231–123n) based on the parity data. If the detected error is within the correction range, the error correction code (ECC) unit (1214) can correct the detected error.

[0101] The memory interface unit (1215) can provide control signals, such as commands and addresses, to non-volatile memory devices (1231 to 123n) under the control of the control unit (1212). The memory interface unit (1215) can also exchange data with the non-volatile memory devices (1231 to 123n) under the control of the control unit (1212). For example, the memory interface unit (1215) can provide data stored in the buffer memory device (1220) to the non-volatile memory devices (1231 to 123n), or provide data read from the non-volatile memory devices (1231 to 123n) to the buffer memory device (1220).

[0102] The buffer memory device (1220) can temporarily store data to be stored in the non-volatile memory devices (1231–123n). Additionally, the buffer memory device (1220) can temporarily store data read from the non-volatile memory devices (1231–123n). Data temporarily stored in the buffer memory device (1220) can be transferred to the host device (1100) or the non-volatile memory devices (1231–123n) under the control of the controller (1210).

[0103] Non-volatile memory devices (1231–123n) can be used as storage media for the SSD (1200). Each of the non-volatile memory devices (1231–123n) can be connected to the controller (1210) through multiple channels (CH1–CHn). One or more non-volatile memory devices can be connected to a single channel. Non-volatile memory devices connected to a single channel can be connected to the same signal bus and data bus.

[0104] The power supply (1240) can provide power (PWR) input through the power connector (1260) to the SSD (1200) background. The power supply (1240) may include an auxiliary power supply (1241). The auxiliary power supply (1241) can supply power so that the SSD (1200) can shut down normally in the event of a sudden power off. The auxiliary power supply (1241) may include large capacitors.

[0105] The signal connector (1250) can be configured with various types of connectors depending on the interface method between the host device (1100) and the SSD (1200).

[0106] The power connector (1260) can be configured with various types of connectors depending on the power supply method of the host device (1100).

[0107] FIG. 9 is a diagram illustrating an exemplary data processing system including a memory system according to an embodiment of the present invention. Referring to FIG. 9, the data processing system (2000) may include a host device (2100) and a memory system (2200).

[0108] The host device (2100) may be configured in the form of a board, such as a printed circuit board. Although not illustrated, the host device (2100) may include background function blocks to perform the functions of the host device.

[0109] The host device (2100) may include a connection terminal (2110), such as a socket, slot, or connector. The memory system (2200) may be mounted to the connection terminal (2110).

[0110] The memory system (2200) may be configured in the form of a substrate such as a printed circuit board. The memory system (2200) may be referred to as a memory module or a memory card. The memory system (2200) may include a controller (2210), a buffer memory device (2220), non-volatile memory devices (2231–2232), a power management integrated circuit (PMIC) (2240), and a connection terminal (2250).

[0111] The controller (2210) can control the general operations of the memory system (2200). The controller (2210) can be configured in the same way as the controller (1210) shown in FIG. 8.

[0112] The buffer memory device (2220) can temporarily store data to be stored in the non-volatile memory devices (2231–2232). Additionally, the buffer memory device (2220) can temporarily store data read from the non-volatile memory devices (2231–2232). Data temporarily stored in the buffer memory device (2220) can be transferred to the host device (2100) or the non-volatile memory devices (2231–2232) under the control of the controller (2210).

[0113] Non-volatile memory devices (2231–2232) can be used as storage media for the memory system (2200).

[0114] The PMIC (2240) can provide power input through the connection terminal (2250) to the background of the memory system (2200). The PMIC (2240) can manage the power of the memory system (2200) under the control of the controller (2210).

[0115] The connection terminal (2250) can be connected to the connection terminal (2110) of the host device. Through the connection terminal (2250), signals such as commands, addresses, data, and power can be transmitted between the host device (2100) and the memory system (2200). The connection terminal (2250) can be configured in various forms depending on the interface method between the host device (2100) and the memory system (2200). The connection terminal (2250) can be placed on one side of the memory system (2200).

[0116] FIG. 10 is a drawing illustrating an exemplary data processing system including a memory system according to an embodiment of the present invention. Referring to FIG. 10, the data processing system (3000) may include a host device (3100) and a memory system (3200).

[0117] The host device (3100) may be configured in the form of a board, such as a printed circuit board. Although not illustrated, the host device (3100) may include background function blocks to perform the functions of the host device.

[0118] The memory system (3200) may be configured in the form of a surface-mount package. The memory system (3200) may be mounted to a host device (3100) via a solder ball (3250). The memory system (3200) may include a controller (3210), a buffer memory device (3220), and a non-volatile memory device (3230).

[0119] The controller (3210) can control the general operations of the memory system (3200). The controller (3210) can be configured to be the same as the controller (1210) shown in FIG. 8 or the controller (2210) shown in FIG. 9.

[0120] The buffer memory device (3220) can temporarily store data to be stored in the non-volatile memory device (3230). Additionally, the buffer memory device (3220) can temporarily store data read from the non-volatile memory devices (3230). Data temporarily stored in the buffer memory device (3220) can be transferred to the host device (3100) or the non-volatile memory device (3230) under the control of the controller (3210).

[0121] A non-volatile memory device (3230) can be used as a storage medium for a memory system (3200).

[0122] FIG. 11 is a diagram illustrating an exemplary network system including a memory system according to an embodiment of the present invention. Referring to FIG. 11, the network system (4000) may include a server system (4300) and a plurality of client systems (4410 to 4430) connected through a network (4500).

[0123] The server system (4300) can provide data in response to requests from multiple client systems (4410 to 4430). For example, the server system (4300) can store data provided by multiple client systems (4410 to 4430). As another example, the server system (4300) can provide data to multiple client systems (4410 to 4430).

[0124] The server system (4300) may include a host device (4100) and a memory system (4200). The memory system (4200) may be composed of the memory system (200) of FIG. 1, the SSD (1200) of FIG. 8, the memory system (2200) of FIG. 9, and the memory system (3200) of FIG. 10.

[0125] FIG. 12 is a block diagram illustrating, by way of example, a non-volatile memory device included in a memory system according to an embodiment of the present invention. Referring to FIG. 12, the non-volatile memory device (300) may include a memory cell array (310), a row decoder (320), a data read / write block (330), a column decoder (340), a voltage generator (350), and control logic (360).

[0126] The memory cell array (310) may include memory cells (MC) arranged in areas where word lines (WL1~WLm) and bit lines (BL1~BLn) intersect each other.

[0127] The row decoder (320) can be connected to the memory cell array (310) through the word lines (WL1–WLm). The row decoder (320) can operate under the control of the control logic (360). The row decoder (320) can decode an address provided from an external device (not shown). The row decoder (320) can select and drive the word lines (WL1–WLm) based on the decoding result. For example, the row decoder (320) can provide a word line voltage provided from the voltage generator (350) to the word lines (WL1–WLm).

[0128] The data read / write block (330) can be connected to the memory cell array (310) through bit lines (BL1 to BLn). The data read / write block (330) may include read / write circuits (RW1 to RWn) corresponding to each of the bit lines (BL1 to BLn). The data read / write block (330) can operate under the control of the control logic (360). Depending on the operation mode, the data read / write block (330) can operate as a write driver or as a sense amplifier. For example, the data read / write block (330) can operate as a write driver that stores data provided from an external device in the memory cell array (310) during a write operation. As another example, the data read / write block (330) can operate as a sense amplifier that reads data from the memory cell array (310) during a read operation.

[0129] The column decoder (340) can operate under the control of the control logic (360). The column decoder (340) can decode an address provided from an external device. Based on the decoding result, the column decoder (340) can connect the read / write circuits (RW1~RWn) of the data read / write block (330) corresponding to each of the bit lines (BL1~BLn) with the data input / output line (or data input / output buffer).

[0130] The voltage generator (350) can generate a voltage used for background operation of the non-volatile memory device (300). The voltages generated by the voltage generator (350) can be applied to the memory cells of the memory cell array (310). For example, a program voltage generated during a program operation can be applied to the word lines of the memory cells where the program operation is to be performed. As another example, an erase voltage generated during an erase operation can be applied to the well regions of the memory cells where the erase operation is to be performed. As another example, a read voltage generated during a read operation can be applied to the word lines of the memory cells where the read operation is to be performed.

[0131] The control logic (360) can control the general operations of the non-volatile memory device (300) based on a control signal provided from an external device. For example, the control logic (360) can control the read, write, and erase operations of the non-volatile memory device (300).

[0132] A person skilled in the art to which the present invention pertains should understand that the present invention may be implemented in other specific forms without altering its technical concept or essential features, and therefore the embodiments described above should be understood as illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims set forth below rather than by the detailed description above, and all modifications or variations derived from the meaning and scope of the claims and equivalent concepts should be interpreted as being included within the scope of the present invention. Explanation of the symbols

[0133] 10: Data Processing System 100: Host device 110: Logical Area Management Unit 120: Processor 130: Host memory buffer 135: Light Status Table 200: Memory System 210: Controller 211: Light Unit 212: Recovery Unit 220: Storage media

Claims

Claim 1 A memory system comprising: a storage medium including physical regions; and a controller configured to map logical regions configured by a host device to the physical regions and to perform a write operation on the physical region mapped to the target logical region in response to a write request for the target logical region, wherein the controller is further configured to update a write state corresponding to the target logical region in a write state table in response to the write request, and the controller is further configured to update all write states in the write state table to a first state when it is determined that there was no abnormal termination immediately before the boot operation. Claim 2 A memory system according to claim 1, wherein the controller is configured to update the light state by changing the light state from the first state to the second state when the light state is determined to be the first state. Claim 3 A memory system according to paragraph 2, wherein the controller is configured to update the light state by maintaining the light state in the second state when it is determined that the light state is the second state. Claim 4 A memory system according to paragraph 2, wherein the controller is further configured to identify a logical region in which the light state is the second state within the light state table when it is determined that there was an abnormal termination immediately before the boot operation, and to perform a recovery operation for a physical region mapped to the identified logical region. Claim 5 A memory system according to claim 4, wherein the controller is further configured to omit a recovery operation for a physical area mapped to a logical area in which the light state within the light state table is the first state when it is determined that there was an abnormal termination immediately before the boot operation. Claim 6 A memory system according to paragraph 4, wherein the controller is further configured to change the light state corresponding to the identified logical region from the second state to the first state after performing the recovery operation. Claim 7 A memory system according to claim 4, wherein the controller is further configured to validate the validity of the light state table based on the validity verification information of the light state table when it is determined that there was an abnormal termination immediately prior to the boot operation. Claim 8 In claim 7, the memory system is further configured such that the controller performs a recovery operation for one or more open physical regions when it is determined that the light state table is invalid. Claim 9 In claim 1, the memory system is further configured such that the controller manages the light status table in a host memory buffer included in the host device. Claim 10 A memory system comprising: a storage medium including physical regions; and a controller configured to receive a write request from a host device, wherein the controller is further configured to change the first state to a second state when the write state corresponding to the target physical region of the write request in a write state table is the first state, and maintain the second state when the write state is the second state, and further configured to update all write states in the write state table to the first state when it is determined that there was no abnormal termination immediately before the boot operation. Claim 11 A memory system according to claim 10, wherein the controller is further configured to omit a recovery operation for a physical region in which the light state in the light state table is the first state when it is determined that there was an abnormal termination immediately before the boot operation. Claim 12 A memory system according to claim 10, wherein the controller is further configured to perform a recovery operation for a physical region in which the light state in the light state table is the second state when it is determined that there was an abnormal termination immediately before the boot operation. Claim 13 In paragraph 12, the memory system is further configured such that the controller changes the light state corresponding to the physical region where the recovery operation was performed from the second state to the first state. Claim 14 In paragraph 10, the above controller is further configured to generate validation information for the light state table whenever the light state table is updated, in a memory system. Claim 15 A memory system according to claim 14, wherein the controller is further configured to validate the validity of the light state table based on the validation information when it is determined that there was an abnormal termination immediately prior to the boot operation. Claim 16 In paragraph 10, the above controller is further configured to manage the light state table in a host memory buffer included in the host device, a memory system. Claim 17 A controller of a memory system comprising: a write unit configured to map logical regions configured by a host device to physical regions and to perform a write operation on a physical region mapped to the target logical region in response to a write request for a target logical region; and a recovery unit configured to omit a recovery operation for a physical region mapped to a logical region in a first state of write status and to perform a recovery operation for a physical region mapped to a logical region in a second state of write status when it is determined that there was an abnormal termination immediately before the boot operation, wherein the recovery unit is further configured to update all managed write states to the first state when it is determined that there was no abnormal termination immediately before the boot operation. Claim 18 A controller of a memory system, wherein the light unit is further configured to change the light state corresponding to the target logic region to the second state when it is determined that the light state corresponding to the target logic region is the first state in response to the light request. Claim 19 A controller of a memory system, wherein, in paragraph 18, the light unit is further configured to maintain the second state when it is determined that the light state corresponding to the target logic region is the second state in response to the light request. Claim 20 A controller of a memory system, wherein, in paragraph 17, the recovery unit is further configured to change the light state from the second state to the first state after performing the recovery operation.

Citation Information

Patent Citations

  • A STT-MRAM-based solid-state memory device power-off recovery method

    CN109284070A

  • Nonvolatile memory device and program method thereof

    KR1020130128694A

  • System and Method for High Performance and Low Cost Flash Translation Layer

    US20140337560A1

  • Power-off recovery management for sector based flash media managers

    US6170066B1