Memory device allocating unit partition region and operating method thereof

By dividing memory blocks into partitioned regions and unit partitioned regions, and adopting the partitioned namespace standard, sequential storage and mapping of data are achieved, which solves the problem of reduced lifespan of non-volatile memory devices during random access and improves the efficiency and reliability of the storage device.

CN118897650APending Publication Date: 2024-11-05SAMSUNG ELECTRONICS CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410538790.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-05-04
Filing Date
2024-04-30
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

Existing non-volatile memory devices suffer from reduced memory lifespan during random access, especially due to increased complexity and resource consumption caused by frequent garbage collection.

Method used

By dividing memory blocks into partitioned regions and unit partitioned regions, and adopting the Partition Namespace (ZNS) standard, sequential storage and mapping of data are achieved, reducing direct management of memory blocks, dynamically adjusting the size or number of partition namespaces, and reducing resource consumption and processing time.

Benefits of technology

It effectively extends the lifespan of storage devices, reduces resource consumption and processing time, simplifies the management of memory blocks, and improves the efficiency and reliability of data storage.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118897650A_ABST
    Figure CN118897650A_ABST
Patent Text Reader

Abstract

A method of operating a storage device in communication with a host device includes allocating a plurality of unit partition regions, each unit partition region having a unit partition size, each of a plurality of unit partition regions is mapped to at least one corresponding memory block of a plurality of memory blocks in a storage device and supports a sequential write in compliance with a partition namespace (ZNS) standard, receiving a first request from a host device indicating an allocation operation of a first partition namespace in compliance with the ZNS standard, and based on the first request, allocating the first partition namespace to the storage device. A first mapping ratio is determined by dividing a first partition size of the first partition namespace by the unit partition size, and the first mapping ratio is stored in a partition metadata table of the storage device.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the priority of Korean Patent Application No. 10-2023-0058676 filed in the Korean Intellectual Property Office on May 4, 2023, the disclosure of which is incorporated herein by reference in its entirety. Technical Field

[0003] The memory device stores data in response to a write request, and outputs the data stored therein in response to a read request. For example, the memory device is classified into a volatile memory device that loses the data stored therein when the power is turned off, such as a dynamic random access memory (DRAM) device or a static RAM (SRAM) device, or a nonvolatile memory device that retains the data stored therein even when the power is turned off, such as a flash memory device, a phase change RAM (PRAM), a magnetic RAM (MRAM), or a resistive RAM (RRAM). Background Art

[0004] Typically, nonvolatile memory devices may rely on random access to store data, which is accompanied by frequent garbage collection of the entire area, resulting in a reduced lifespan of the memory device. Summary of the invention

[0005] Embodiments of the present disclosure provide a storage device for allocating a unit zone region and an operating method thereof.

[0006] According to an embodiment, a method for operating a storage device that communicates with a host device includes: allocating a plurality of unit partition areas, each unit partition area having a unit partition size, each unit partition area among the plurality of unit partition areas being mapped to at least one corresponding memory block among a plurality of memory blocks in the storage device and supporting sequential writes that comply with a zoned namespace (ZNS) standard, receiving a first request indicating an allocation operation of a first partition namespace that complies with the ZNS standard from the host device, determining a first mapping ratio based on the first request by dividing a first partition size of the first partition namespace by the unit partition size, and storing the first mapping ratio in a partition metadata table of the storage device.

[0007] According to an embodiment, a method for operating a storage device that communicates with a host device includes: allocating multiple unit partition areas, each unit partition area having a corresponding unit partition area (UZR) index value, receiving a request indicating an allocation operation of a first partition namespace and a second partition namespace that conforms to a partition namespace (ZNS) standard, determining a first maximum UZR index value for the first partition namespace based on the request, wherein the first maximum UZR index value corresponds to the last unit partition area among the multiple unit partition areas corresponding to the first partition namespace, determining a second maximum UZR index value for the second partition namespace based on the request, wherein the second maximum UZR index value corresponds to the last unit partition area among the multiple unit partition areas corresponding to the second partition namespace, and storing the first maximum UZR index value and the second maximum UZR index value in a partition metadata table of the storage device.

[0008] According to an embodiment, a storage device includes: a command manager that receives a first request indicating an allocation operation of a first partition namespace that complies with a partition namespace (ZNS) standard from a host device, and generates a second request indicating the allocation operation; a partition manager that allocates a plurality of unit partition areas, each of which has a unit partition size and a corresponding unit partition area (UZR) index value; a partition metadata table; and a non-volatile memory device that includes a plurality of memory blocks corresponding to the plurality of unit partition areas. The partition manager determines a first mapping ratio by dividing a first partition size of the first partition namespace by the unit partition size based on the second request, stores the first mapping ratio in the partition metadata table, determines a first maximum UZR index value based on the second request, the first maximum UZR index value corresponds to the last unit partition area among the plurality of unit partition areas corresponding to the first partition namespace, and stores the first maximum UZR index value in the partition metadata table. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other advantages and features of the present disclosure will become more apparent by describing in detail embodiments of the present disclosure with reference to the accompanying drawings.

[0010] Figure 1 is a block diagram of an electronic device according to some embodiments of the present disclosure.

[0011] Figure 2 The operation of some embodiments of the present disclosure is described Figure 1 A schematic diagram of the method of an electronic device.

[0012] Figure 3 is a schematic diagram describing how a conventional electronic device manages a namespace.

[0013] Figure 4 is a schematic diagram describing the management of namespaces according to some embodiments of the present disclosure.

[0014] Figure 5 is a schematic diagram describing the states of partitioned areas according to some embodiments of the present disclosure.

[0015] Figure 6 is a schematic diagram describing sequential writing of partitioned areas according to some embodiments of the present disclosure.

[0016] Figure 7 is a schematic diagram describing how to map partition areas and unit partition areas according to some embodiments of the present disclosure.

[0017] Figure 8 is a schematic diagram describing how to map partition areas and unit partition areas according to some embodiments of the present disclosure.

[0018] Fig. 9 is a flowchart describing a method of operating an electronic device according to some embodiments of the present disclosure.

[0019] Fig.10 Describes some embodiments of the present disclosure Fig. 9 Schematic diagram of the reallocation operation.

[0020] Fig.11 Describes some embodiments of the present disclosure Fig. 9 Schematic diagram of the reallocation operation.

[0021] Fig.12 is a flowchart describing a method of operating a storage device according to some embodiments of the present disclosure.

[0022] Fig.13 is a flowchart describing a method of operating a storage device according to some embodiments of the present disclosure.

[0023] Fig.14 is a flowchart describing a method of operating a storage device according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0024] Hereinafter, embodiments of the present disclosure will be described in detail and clearly so that those skilled in the art can easily perform the embodiments of the present disclosure.

[0025] Figure 1 is a block diagram of an electronic device according to some embodiments of the present disclosure. Figure 1, the electronic device 1000 includes a host device 1100 and a storage device 1200. The electronic device 1000 may be a computing system configured to process various information, such as a personal computer (PC), a notebook computer, a laptop, a server, a workstation, a tablet PC, a smart phone, a digital camera, and a black box.

[0026] The host device 1100 may control the overall operation of the electronic device 1000. The host device 1100 may include an application, a file system, and a device driver. The application may generate a user request. The file system may generate a system request for generating, changing, or deleting a file based on the user request. Based on the system request, the device driver may store data in the storage device 1200, may read data stored in the storage device 1200, or may manage data stored in the storage device 1200. The device driver may support the partition name space (ZNS) standard of NVMe (Non-Volatile Memory Express).

[0027] The ZNS standard may be a protocol that supports sequential writes in a partitioned region ZR. Conventional nonvolatile memory devices may store data depending on random access. For example, data corresponding to sequential logical block addresses may be stored in non-physical sequential memory cells or memory blocks. Random access is accompanied by frequent garbage collection of the entire region, resulting in a reduced life of the storage device 1200.

[0028] To prevent the above problems, the memory block of the nonvolatile memory device can be divided into partition areas, and the data can be stored sequentially in the partition areas or in subcomponents of the partition areas (e.g., unit partition areas). Storing data sequentially can mean storing data corresponding to sequential logical block addresses at sequential physical block addresses.

[0029] The host device 1100 can manage multiple partition areas ZR1 to ZRN. Here, N is an arbitrary natural number. The partition area may also be referred to as a "host partition area" that can be recognized by the host device 1100. The partition area may correspond to a logically sequential logical block address. The partition area may be a subcomponent of a namespace (e.g., a partition namespace (ZNS)). In order to prevent conflicts between objects with the same name, a namespace may be a logical area (e.g., in a related namespace) that stores a mapping relationship between an object and an identifier belonging thereto.

[0030] The storage device 1200 may store data under the control of the host device 1100. The storage device 1200 may support the ZNS standard of NVMe. The storage device 1200 may support a unit partition region (UZR) whose size is smaller than the size of a partition region of the host device 1100. For example, the storage device 1200 may allocate a plurality of unit partition regions UZR1 to UZRM. Here, M is an arbitrary natural number.

[0031] Each of the plurality of unit partition areas UZR1 to UZRM may be mapped to at least one memory block BLK. The unit partition area may conceptually refer to some of the memory blocks BLK of the nonvolatile memory device 1220. In the unit partition area, data corresponding to sequential logical block addresses may be stored in sequential memory blocks or in sequential physical pages of memory blocks. That is, the unit partition area may support sequential writes that comply with the ZNS standard. The storage device 1200 may manage a mapping relationship between a partition area and a unit partition area. The host device 1100 may identify the partition area, but it may be difficult to identify the unit partition area.

[0032] Each of the plurality of unit partition areas UZR1 to UZRM may have a unit partition size. The unit partition size may refer to the size of a physical space corresponding to one unit partition area. For example, the unit partition size may refer to the number of memory blocks BLK corresponding to one unit partition area or the number of physical pages corresponding to one unit partition area (e.g., the number of subcomponents of a memory block BLK).

[0033] In some embodiments, the unit partition size corresponds to one memory block, a first number of memory blocks according to an operation policy of the storage device 1200 , or a second number of memory blocks determined by the host device 1100 .

[0034] Each of the plurality of unit partition areas UZR1 to UZRM may have a corresponding UZR index value. The UZR index value may be used to identify a corresponding unit partition area among the plurality of unit partition areas UZR1 to UZRM. Each of the plurality of unit partition areas UZR1 to UZRM may have a unique UZR index value.

[0035] The storage device 1200 may include a storage controller 1210 and a nonvolatile memory device 1220. Under the control of the host device 1100 or depending on an internal policy, the storage controller 1210 may store data in the nonvolatile memory device 1220 or may read data stored in the nonvolatile memory device 1220. For example, based on a command CMD indicating an operation to be performed and an address ADD indicating a location of data, the storage controller 1210 may store data in the nonvolatile memory device 1220 or may read data stored in the nonvolatile memory device 1220.

[0036] The storage controller 1210 may include a command manager 1211, a partition manager 1212, and a partition metadata table 1213. In response to a command received from the host device 1100, the command manager 1211 may perform an operation corresponding to the request or may control the partition manager 1212.

[0037] The partition manager 1212 may allocate a plurality of unit partition areas UZR1 to UZRM. The partition manager 1212 may manage a mapping relationship between the partition areas ZR1 to ZRN and the unit partition areas UZR1 to UZRM. The partition manager 1212 may generate metadata of the partition areas ZR1 to ZRN and the unit partition areas UZR1 to UZRM. The partition manager 1212 may store the metadata in a partition metadata table 1213. The metadata may include the size of a namespace clearly stated by the host device 1100 (e.g., the size of a partition namespace), the size of a partition area, the status of a unit partition area, a write pointer WP of a unit partition area, and the like. Reference will be made to Figure 2 Describes the metadata in detail.

[0038] The partition metadata table 1213 stores metadata under the control of the partition manager 1212 .

[0039] In some embodiments, the command manager 1211, the partition manager 1212, and the partition metadata table 1213 are implemented by firmware. For example, the storage controller 1210 may include a processor and a volatile memory device. The non-volatile memory device 1220 may store instructions corresponding to the modules implemented by the firmware. The processor may load the instructions stored in the non-volatile memory device 1220 to the volatile memory device. The processor may execute the instructions loaded to the volatile memory device to allow the storage device 1200 to perform the functions of the module.

[0040] The nonvolatile memory device 1220 may store data under the control of the memory controller 1210. The nonvolatile memory device 1220 may include a memory block BLK corresponding to a plurality of unit partition regions UZR1 to UZRM. The memory block BLK may include a plurality of memory cells. The memory cells may store data.

[0041] In some embodiments, the nonvolatile memory device 1220 is a NAND flash memory, but the present disclosure is not limited thereto. The nonvolatile memory device 1220 may be implemented with one of various memory devices that retain data stored therein even when power is turned off, such as a phase change random access memory (PRAM), a magnetic random access memory (MRAM), a resistive random access memory (RRAM), and a ferroelectric random access memory (FRAM).

[0042] Figure 2 The operation of some embodiments of the present disclosure is described Figure 1 A schematic diagram of the electronic device method. Figure 1 and Figure 2 , the electronic device 1000 includes a host device 1100 , a command manager 1211 , a partition manager 1212 , a partition metadata table 1213 , and a non-volatile memory device 1220 .

[0043] The host device 1100 may support a multi-ZNS environment. A multi-ZNS environment may refer to an environment that supports at least two different partition namespaces. For example, the host device 1100 may manage a first partition namespace ZNS1 and a second partition namespace ZNS2. The entire size of the namespace, the size of the partition area as a subcomponent, and the like may be set differently for each namespace. The first partition namespace ZNS1 may include partition areas ZR11 to ZR14. Partition areas ZR11 to ZR14 may be logically sequential. Partition area ZR11 may be referred to as the "first partition area" of the first partition namespace ZNS1. Partition area ZR14 may be referred to as the "last partition area" of the first partition namespace ZNS1. The second partition namespace ZNS2 may include partition areas ZR21 to ZR24. Partition areas ZR21 to ZR24 may be logically sequential. Partition area ZR21 may be referred to as the "first partition area" of the second partition namespace ZNS2. Partition area ZR24 may be referred to as the "last partition area" of the second partition namespace ZNS2. The partition region ZR14 which is the last partition region among the partition regions ZR11 to ZR14 of the first partition namespace ZNS1 may be logically sequential with the partition region ZR21 which is the first partition region among the partition regions ZR21 to ZR24 of the second partition namespace ZNS2.

[0044] For better understanding of the present disclosure, 2 namespaces and 8 regions are shown, but the present disclosure is not limited thereto. The number of namespaces supported by the host device 1100 may increase or decrease, and the number of regions in a namespace may increase or decrease.

[0045] The partition metadata table 1213 may include a namespace size table 1213a, a partition size table 1213b, and a partition status table 1213c. The metadata may include the maximum UZR index value of the partition namespace, the mapping ratio of the partition namespace, the status of the unit partition area, the write pointer of the unit partition area, etc.

[0046] The namespace size table 1213a may store metadata about the entire size of the partition namespace. The namespace size table 1213a may store a set of ZNS index values ​​and maximum UZR index values. The ZNS index value may be a unique value for identifying a partition namespace corresponding to the ZNS index value among a plurality of partition namespaces managed by the host device 1100. The maximum UZR index value may be a UZR index value of the last unit partition area (or the unit partition area with the largest logical number) mapped to the unit partition area of ​​the corresponding partition namespace. The maximum UZR index value may be used to identify the entire size of the partition namespace. The entire size of the partition namespace may correspond to the maximum UZR index value, or may correspond to the difference between the maximum UZR index value and the maximum UZR index value of the previous partition namespace.

[0047] The partition size table 1213b may store metadata about the entire size of the partition area of ​​the partition namespace. The partition size may indicate the logical size of the partition area included in the partition namespace. The partition size may be determined relative to the unit partition area and may correspond to a mapping ratio. The partition size table 1213b may store a set of ZNS index values ​​and mapping ratios. The mapping ratio may be a value obtained by dividing the size of the partition area of ​​the relevant partition namespace by the size of the unit partition area. For example, when the mapping ratio is 2, the size of the partition area may be twice the size of the unit partition area. As another example, when the mapping ratio is 1, the size of the partition area may be the same as the size of the unit partition area.

[0048] The partition status table 1213c may store metadata about the status and write pointer of each unit partition area, which complies with the ZNS standard. The status may indicate the partition status of the unit partition area that complies with the ZNS standard. Figure 5 An example of a partition state is described in detail. The write pointer may point to the last logical block address in the written logical block addresses of the partition area and may indicate where the next write operation starts. Figure 6The write pointer is described in detail. The partition status table 1213c may store a set of UZR index values, status, and write pointers.

[0049] The metadata of the partition state table 1213c may be updated based on a subsequent user operation (eg, a state conversion operation, a read operation, a write operation, or a metadata read operation) after the allocation operation. Fig. 9 Describe the user action in detail.

[0050] In some embodiments, the metadata of the unit partition area stored in the partition status table 1213c is used to determine the metadata of the partition area. For example, the partition area ZR11 of the first partition namespace ZNS1 may correspond to the unit partition area UZR1 and the unit partition area UZR2. The state of the partition area ZR11 may be determined based on the state of the unit partition area UZR1 and the state of the unit partition area UZR2. The write pointer of the partition area ZR11 may correspond to the sum of the write pointer of the unit partition area UZR1 and the write pointer of the unit partition area UZR2. Figure 7 , Figure 8 and Fig.14 Describe in detail how metadata is collected.

[0051] The nonvolatile memory device 1220 may include a plurality of memory blocks BLK. The plurality of memory blocks BLK may correspond to a plurality of unit partition areas UZR1 to UZRM. Some of the plurality of unit partition areas UZR1 to UZRM may have a mapping relationship with the partition areas ZR11 to ZR14 of the first partition namespace ZNS1. Other unit partition areas of the plurality of unit partition areas UZR1 to UZRM may have a mapping relationship with the partition areas ZR21 to ZR24 of the second partition namespace ZNS2. Each of the plurality of unit partition areas UZR1 to UZRM may have a unit partition size. Each of the plurality of unit partition areas UZR1 to UZRM may have a UZR index value.

[0052] Next, a method of operating the electronic device 1000 of the present disclosure will be described.

[0053] In the first operation ①, the host device 1100 may provide a request RQ for an allocation operation to the command manager 1211. The allocation operation may include an allocation operation of a first partition namespace ZNS1 and an allocation operation of a second partition namespace ZNS2 that conform to the ZNS standard. The request RQ may include information about a first partition size of the first partition namespace ZNS1 and information about a second partition size of the second partition namespace ZNS2. The command manager 1211 may provide a request for partition management to the partition manager 1212 based on the request RQ.

[0054] In the second operation ②, the partition manager 1212 may perform an allocation operation based on a request from the command manager 1211. The allocation operation may refer to an operation of generating metadata corresponding to a mapping relationship between a partition area and a unit partition area and storing the metadata in a partition metadata table 1213. The size of a unit partition area or the size of a partition area may be set by updating only metadata about a mapping relationship between a partition area and a unit partition area without an allocation, deallocation, or reallocation operation associated with a unit partition area and a memory block BLK. Since the allocation, deallocation, or reallocation operation of a memory block BLK is actually omitted, resource consumption, power consumption, and processing time may be reduced.

[0055] For example, the partition manager 1212 determines the first mapping ratio by dividing the first partition size of the first partition namespace ZNS1 by the unit partition size. The partition area of ​​the first partition namespace ZNS1 may correspond to two unit partition areas. The first mapping ratio may be 2. The partition manager 1212 may store the first mapping ratio of the first partition namespace ZNS1 in the partition size table 1213b.

[0056] As described above, the partition manager 1212 may determine the second mapping ratio by dividing the second partition size of the second partition namespace ZNS2 by the unit partition size. The partition area of ​​the second partition namespace ZNS2 may correspond to one unit partition area. The second mapping ratio may be 1. The partition manager 1212 may store the second mapping ratio of the second partition namespace ZNS2 in the partition size table 1213b.

[0057] As another example, the partition manager 1212 may determine a first maximum UZR index value of the first partition namespace ZNS1 based on a request from the command manager 1211. The first partition namespace ZNS1 may include 4 partition areas ZR11 to ZR14. The first mapping ratio may be 2. The first partition namespace ZNS1 may correspond to 8 unit partition areas UZR1 to UZR8. The first maximum UZR index value may correspond to UZR8, which is the last unit partition area among the unit partition areas UZR1 to UZR8 corresponding to the first partition namespace ZNS1 among the plurality of unit partition areas UZR1 to UZRM. The partition manager 1212 may store the first maximum UZR index value in the namespace size table 1213a.

[0058] As described above, the partition manager 1212 may determine the second maximum UZR index value of the second partition namespace ZNS2 based on the request from the command manager 1211. The first maximum UZR index value of the first partition namespace ZNS1, which is the immediately preceding partition namespace, may correspond to UZR8. The second partition namespace ZNS2 may include 4 partition areas ZR21 to ZR24. The second mapping ratio may be 1. The second partition namespace ZNS2 may correspond to 4 unit partition areas UZR9 to UZR12. The second maximum UZR index value may correspond to UZR12, which is the last unit partition area among the unit partition areas UZR9 to UZR12 corresponding to the second partition namespace ZNS2 among the multiple unit partition areas UZR1 to UZRM. The partition manager 1212 may store the second maximum UZR index value in the namespace size table 1213a.

[0059] Figure 3 is a schematic diagram describing how a conventional electronic device manages a namespace. Figure 3 , the conventional electronic device ED includes a host device and a storage device. The storage device may include a nonvolatile memory device. The nonvolatile memory device may include memory blocks BLK1 to BLK16.

[0060] The host device may manage a first partition namespace ZNS1 to a third partition namespace ZNS3. The first partition namespace ZNS1 may include partition regions ZR11 and ZR12. Partition region ZR11 may be mapped to sequential memory blocks BLK1, BLK2, and BLK3. Memory block BLK1 may be directly mapped to partition ZR11. Memory block BLK2 may be mapped to memory block BLK1. Memory block BLK3 may be mapped to memory block BLK2. As described above, partition region ZR12 may be mapped to sequential memory blocks BLK4, BLK5, and BLK6.

[0061] The second partition namespace ZNS2 may include partition regions ZR21, ZR22, ZR23, and ZR24. Partition region ZR21 may be mapped to memory block BLK7. Partition region ZR22 may be mapped to memory block BLK8. Partition region ZR23 may be mapped to memory block BLK9. Partition region ZR24 may be mapped to memory block BLK10.

[0062] The third partition namespace ZNS3 may include partition regions ZR31, ZR32, and ZR33. Partition region ZR31 may be mapped to sequential memory blocks BLK11 and BLK12. Partition region ZR32 may be mapped to sequential memory blocks BLK13 and BLK14. Partition region ZR33 may be mapped to sequential memory blocks BLK15 and BLK16.

[0063] As described above, the conventional electronic device ED can directly map the partition area of ​​the host device to the memory block. In the conventional electronic device ED, whenever the size of the partition namespace or the partition area or the number of the partition namespace or the partition area changes, it may be necessary to change the mapping relationship associated with all the memory blocks of the non-volatile memory device; in this case, the amount of memory access (for example, for generating or changing the link relationship of the memory block) may increase.

[0064] In addition, when the partition size of the partition area is set differently for each partition namespace, the number of memory blocks mapped to the partition area of ​​each partition namespace may be determined differently. In this case, the mapping relationship may be complicated.

[0065] Furthermore, when an exceptional case occurs in which a partition area is mapped to memory blocks whose number is greater than an expected value, a garbage collection (GC) target may be determined by sequentially retrieving linked memory blocks; in this case, the burden of retrieving physical link relationships may increase.

[0066] Figure 4 is a schematic diagram describing the management of namespaces according to some embodiments of the present disclosure. Figure 4 , the electronic device 1000 includes a host device 1100 and a storage device 1200. The storage device 1200 may include a non-volatile memory device 1220. The non-volatile memory device 1220 may include memory blocks BLK1 to BLK16. The storage device 1200 may allocate unit partition areas UZR1 to UZR16 to the memory blocks BLK1 to BLK16, respectively. In order to better understand the present disclosure, a description will be given in which one unit partition area is allocated to one memory block. However, the present disclosure is not limited thereto. For example, one unit partition area may be allocated to a plurality of memory blocks.

[0067] The host device 1100 can manage the first partition namespace ZNS1 to the third partition namespace ZNS3. The first partition namespace ZNS1 may include partition areas ZR11 and ZR12. Partition area ZR11 may be mapped to unit partition areas UZR1, UZR2, and UZR3. Generating a mapping relationship between partition area ZR11 and unit partition areas UZR1, UZR2, and UZR3 may mean generating metadata of partition area ZR11 and unit partition areas UZR1, UZR2, and UZR3. Even if unit partition areas UZR1, UZR2, and UZR3 are newly mapped to partition area ZR11, the mapping relationship between unit partition areas UZR1, UZR2, and UZR3 and memory blocks BLK1 to BLK3 may be maintained without modification. As described above, partition area ZR12 may be mapped to unit partition areas UZR4, UZR5, and UZR6.

[0068] The storage device 1200 may manage the entire size of the first partition namespace ZNS1, the number of partition areas of the first partition namespace ZNS1, and the partition size of the partition area of ​​the first partition namespace ZNS1 based on metadata. For example, the metadata may include a maximum UZR index value corresponding to the last unit partition area (e.g., UZR6) in the unit partition area corresponding to the first partition namespace ZNS1. The storage device 1200 may identify the entire size of the partition namespace based on the maximum UZR index value. The metadata may include a ZR-UZR mapping ratio (e.g., 3) of the first partition namespace ZNS1. The storage device 1200 may identify the partition size of the partition area based on the mapping ratio. The storage device 1200 may identify the number of partition areas in the partition namespace based on the mapping ratio and the maximum UZR index value.

[0069] The second partition namespace ZNS2 may include partition areas ZR21, ZR22, ZR23, and ZR24. Partition area ZR21 may be mapped to unit partition area UZR7. Partition area ZR22 may be mapped to unit partition area UZR8. Partition area ZR23 may be mapped to unit partition area UZR9. Partition area ZR24 may be mapped to unit partition area UZR10.

[0070] The third partition namespace ZNS3 may include partition regions ZR31, ZR32, and ZR33. Partition region ZR31 may be mapped to unit partition regions UZR11 and UZR12. Partition region ZR32 may be mapped to unit partition regions UZR13 and UZR14. Partition region ZR33 may be mapped to unit partition regions UZR15 and UZR16.

[0071] As described above, the electronic device 1000 may pre-allocate the unit partition area before allocating the partition area of ​​the partition namespace. In the electronic device 1000, even if the size of the partition namespace or the partition area or the number of the partition namespace or the partition area changes, the mapping relationship between the unit partition area and the memory block may not be changed, and therefore, memory access (e.g., for generating or changing the link relationship of the memory block) may not be required. Therefore, an electronic device 1000 capable of dynamically changing the size of the partition namespace or the partition area or the number of the partition namespace or the partition area can be provided.

[0072] In addition, when the partition size of the partition area is set differently for each partition namespace, the electronic device 1000 may manage a mapping ratio for each partition namespace. With reference to the mapping ratio, the electronic device 1000 may easily manage partition namespaces of different sizes.

[0073] In addition, when an exception occurs in which a partition area is mapped to a memory block whose number is greater than an expected value, a garbage collection (GC) target can be determined with reference to mapping information of the unit partition area and the memory block. For example, when a unit partition area is mapped to one memory block, the electronic device 1000 can easily determine the garbage collection (GC) target by identifying the unit partition area mapped to two or more memory blocks. Therefore, the burden of retrieving the physical link relationship with the memory block can be reduced.

[0074] Figure 5 is a schematic diagram describing the state of a partitioned area according to some embodiments of the present disclosure. Figure 1 and Figure 5 , Figure 5 The status shown is Figure 1 Each of the partitioned regions ZR1 to ZRN can have a state of either Figure 1 Each of the unit partition areas UZR1 to UZRM can have a state.

[0075] According to some embodiments of the present disclosure, the partition area and the unit partition area managed by the electronic device 1000 have one of the ZSE state, the ZSIO state, the ZSEO state, the ZSC state, the ZSF state, the ZSRO state, and the ZSO state. When the storage device 1200 processes a request received from the host device 1100, the state of the partition area and the unit partition area can be converted.

[0076] The ZSE state, ZSF state, ZSRO state, and ZSO state may be classified as inactive states. The ZSIO state, ZSEO state, and ZSC state may be classified as active states. The ZSIO state and ZSEO state may be classified as open states.

[0077] The ZSE state may indicate an empty state. In the ZSE state, the physical memory block may not yet store data, and the write pointer may point to the lowest logical block address (e.g., the logical block address LBA with the minimum number of logical block addresses corresponding to the partition area or unit partition area). The write pointer of the ZSE state may be valid. The ZSE state may be converted to one of the ZSIO state, the ZSEO state, the ZSC state, and the ZSF state.

[0078] The ZSIO state may indicate an implicit open state. The ZSIO state may be a state implicitly opened by executing a write command received from the host device 1100. In the ZSIO state, the physical memory block may store data corresponding to the write command. The ZSIO state may be converted to one of the ZSE state, the ZSEO state, the ZSC state, and the ZSF state. When the open resources are saturated, the ZSIO state may be converted to the ZSC state even without a close command.

[0079] The ZSEO state may indicate an explicit open state. The ZSEO state may be a state explicitly opened by executing an open command from the host device 1100. In the ZSEO state, the physical memory block may subsequently receive a write command and may then store data corresponding to the write command. The ZSEO state may transition to one of the ZSE state, the ZSC state, and the ZSF state. The ZSEO state may have a higher priority for open resources than the ZSIO state. The ZSEO state may transition to the ZSC state only by a close command.

[0080] The ZSC state may indicate a shutdown state. 1) When there are available active resources in the ZSE state and a set partition descriptor extension command is received, 2) When a shutdown command is received in the ZSIO state or an open resource is saturated, or 3) When a shutdown command is received in the ZSEO state, a partition area or a unit partition area may enter the ZSC state. In the ZSC state, the physical memory block may not store data corresponding to a write command. The ZSC state may transition to one of the ZSE state, the ZSIO state, the ZSEO state, and the ZSF state.

[0081] The ZSF state may indicate a full state. In the ZSE state, the physical memory block may be filled with stored data, and the write pointer may indicate the highest logical block address (e.g., the logical block address LBA having the maximum number of logical block addresses corresponding to the partition area or unit partition area). The write pointer of the ZSF state may be invalid. The ZSF state may be converted to the ZSE state by a reset partition command.

[0082] The ZSRO state may indicate a read-only state. The ZSRO state may correspond to a space in which the host device 1100 continues to use a partition namespace after a portion of the execution of the partition area or the unit partition area stops operating. The ZSRO state may be converted to the ZSO state by an offline partition command.

[0083] The ZSO state may indicate an offline state. The ZSO state may not have a valid write pointer, may not have active resources, and may not have open resources. The ZSO state may not have any other states that it can transition to.

[0084] Figure 6 is a schematic diagram describing sequential writing of partitioned areas according to some embodiments of the present disclosure. Figure 1 and Figure 6 , the storage controller 1210 performs a sequential write operation of data in the nonvolatile memory device 1220. The electronic device 1000 may manage a plurality of partition namespaces ZNS1 to ZNSK. Here, K is an arbitrary natural number.

[0085] Referring to the logical area, the first partition namespace ZNS1 may include partition areas ZR1, ZR2, and ZR3. The mapping ratio of the first partition namespace ZNS1 may be 4. Each of the partition areas ZR1, ZR2, and ZR3 may be mapped to 4 unit partition areas. For example, the partition area ZR1 may be mapped to unit partition areas UZR1 to UZR4. The unit partition area UZR1 may be mapped to logical block addresses LBA1 to LBAy. Here, y is an arbitrary natural number. The logical block addresses LBA1 to LBAy may be in logical order. The range of the unit partition area UZR1 may be from the partition start logical block address ZSLBA to the partition last logical block address ZLLBA. The area start logical block address ZSLBA may be LBA1. The partition last logical block address ZLLBA may be an address obtained by adding the logical block address LBAy to the unit size of the logical block address (for example, it may be the partition start logical block address of the partition area ZR2).

[0086] The storage controller 1210 can perform sequential write operations by using a write pointer. For example, the logical block addresses LBA1 to LBAx of the unit partition area UZR1 can be logical block addresses for storing data (e.g., written logical block addresses). Here, x can be a natural number less than y. The shaded area indicates a logical or physical area where data is stored. The logical block addresses LBAx+1 to LBAy of the unit partition area UZR1 can be logical block addresses where data has not yet been stored (e.g., unwritten logical block addresses). The write pointer can indicate the last logical block address (e.g., LBAx) of the logical block addresses at which the write operation is completed. The data of subsequent write operations can be stored in the physical area corresponding to LBAx+1 (e.g., memory block BLKx+1); after the data is stored, the write pointer can be updated to indicate the next logical block address (e.g., LBAx+1).

[0087] Referring to the physical area, the nonvolatile memory device 1220 may include a plurality of memory blocks corresponding to the unit partition areas UZR1 to UZRM. The unit partition area UZR1 may include memory blocks BLK1 to BLKy. The memory blocks BLK1 to BLKy may be physically sequential (e.g., the memory cells of the memory blocks BLK1 to BLKy may be arranged to be adjacent to each other). The memory blocks BLK1 to BLKy may be identified by physical block addresses PBA1 to PBAy, respectively. The physical block addresses PBA1 to PBAy may correspond to logical block addresses LBA1 to LBAy of the logical area, respectively. The memory blocks BLK1 to BLKx may store data corresponding to the logical block addresses LBA1 to LBAx. The memory blocks BLKx+1 to BLKy may not yet store data and may be used to store data for subsequent write operations.

[0088] In order to better understand the present disclosure, a description is given that one logical block address corresponds to one physical block address and one memory block, but the present disclosure is not limited thereto. The logical block addresses may correspond to sequential sub-blocks, physical pages, or sequential programming units (e.g., units by which memory cells are programmed) of one physical block, respectively, while maintaining the logical sequence feature. The correspondence between the logical block addresses and the memory blocks may be changed and implemented in various ways by a person skilled in the art.

[0089] Figure 7 Schematic diagram for describing how to map partition areas and unit partition areas according to some embodiments of the present disclosure. Figure 7 , metadata associated with a mapping relationship between a partition area and a unit partition area will be described. The unit partition area may include 16 memory blocks BLK1 to BLK16.

[0090] The metadata of the partition area ZR1 may include a mapping ratio, a status, and a write pointer. The mapping ratio of the partition area ZR1 may be 4. The partition area ZR1 may be mapped to 4 unit partition areas UZR1 to UZR4. The status of the partition area ZR1 may be an open state (e.g., a ZSIO state or a ZSEO state). The status of the partition area ZR1 may be determined based on the status of the corresponding unit partition areas UZR1 to UZR4. The write pointer of the partition area ZR1 may be 31. The write pointer of the partition area ZR1 may be the sum of the write pointers of the corresponding unit partition areas UZR1 to UZR4.

[0091] The metadata of the unit partition area UZR1 may include a status and a write pointer. The status of the unit partition area UZR1 may be an open status. For example, the unit partition area UZR1 may have an open status based on an open command or a write command. The write pointer of the unit partition area UZR1 may be 8. For example, the unit partition area UZR1 may be mapped to memory blocks BLK1 to BLK16. Data may exist in memory blocks BLK1 to BLK8. Data may not yet exist in memory blocks BLK9 to BLK16. The write pointer may point to the last memory block in the memory blocks BLK1 to BLK8 storing data. Figure 8 , the shaded blocks indicate memory blocks storing data.

[0092] As described above, the metadata of the unit partition area UZR2 may include a status and a write pointer. The status of the unit partition area UZR2 may be an open state. The write pointer of the unit partition area UZR2 may be 8. The metadata of the unit partition area UZR3 may include a status and a write pointer. The status of the unit partition area UZR3 may be an open state. The write pointer of the unit partition area UZR3 may be 8. The metadata of the unit partition area UZR4 may include a status and a write pointer. The status of the unit partition area UZR4 may be an open state. The write pointer of the unit partition area UZR4 may be 7.

[0093] Referring to the partition area ZR1 and the unit partition areas UZR1 to UZR4 together, since at least one of the unit partition areas UZR1 to UZR4 has an open state, the state of the partition area ZR1 may be determined to be an open state. Since the sum of the write pointers of the unit partition areas UZR1 to UZR4 is 31, the write pointer of the partition area ZR1 may be 31.

[0094] In some embodiments, the sequential characteristics of the unit partition area may be maintained, but the sequential characteristics of the partition area may not be maintained. For example, the storage device may store data in the memory blocks BLK1 to BLK4 of the partition area ZR1, and then may store data in the memory blocks BLK1 to BLK4 of the partition area ZR2, and then may store data in the memory blocks BLK5 to BLK8 of the partition area ZR1. Data may be sequentially stored in the memory blocks corresponding to one unit partition area, but the order of storing data may be random (or non-sequential) between the unit partition areas.

[0095] Figure 8 Schematic diagram for describing how to map partition areas and unit partition areas according to some embodiments of the present disclosure. Figure 8 , metadata associated with a mapping relationship between a partition area and a unit partition area will be described. The unit partition area may include 16 memory blocks BLK1 to BLK16.

[0096] The metadata of the partition area ZR1 may include a mapping ratio, a state, and a write pointer. The mapping ratio of the area ZR1 may be 1. The partition area ZR1 may be mapped to the unit partition area UZR1. The state of the unit partition area ZR1 may be an open state. The write pointer of the partition area ZR1 may be 8.

[0097] As described above, the metadata of the unit partition area ZR2 may include a mapping ratio, a status, and a write pointer. The mapping ratio, status, and write pointer of the partition area ZR2 may be 1, an open state, and 8, respectively. The metadata of the partition area ZR3 may include a mapping ratio, a status, and a write pointer. The mapping ratio, status, and write pointer of the partition area ZR3 may be 1, an open state, and 8, respectively. The metadata of the partition area ZR4 may include a mapping ratio, a status, and a write pointer. The mapping ratio, status, and write pointer of the partition area ZR4 may be 1, an open state, and 7, respectively.

[0098] The metadata of the unit partition area UZR1 may include a status and a write pointer. The status of the unit partition area UZR1 may be an open state. The write pointer of the unit partition area UZR1 may be 8. As described above, the metadata of the unit partition area UZR2 may include a status and a write pointer. The status of the unit partition area UZR2 may be an open state. The write pointer of the unit partition area UZR2 may be 8. The metadata of the unit partition area UZR3 may include a status and a write pointer. The status of the unit partition area UZR3 may be an open state. The write pointer of the unit partition area UZR3 may be 8. The metadata of the unit partition area UZR4 may include a status and a write pointer. The status of the unit partition area UZR4 may be an open state. The write pointer of the unit partition area UZR4 may be 7.

[0099] Referring to the partitioned areas ZR1 to ZR4 and the unit partitioned areas UZR1 to UZR4 together, since the mapping ratio is 1, the partitioned areas ZR1 to ZR4 may be respectively identical to the unit partitioned areas UZR1 to UZR4. In this case, the state of the partitioned areas may be identical to the state of the unit partitioned areas. The write pointer of the partitioned areas may be identical to the write pointer of the unit partitioned areas. That is, the metadata of the partitioned areas having a mapping ratio of 1 may refer to the metadata of the corresponding unit partitioned areas, rather than collecting the metadata of the corresponding unit partitioned areas.

[0100] Fig. 9 is a flowchart describing a method of operating an electronic device according to some embodiments of the present disclosure. Fig. 9 , the electronic device 1000 includes a host device 1100 and a storage device 1200 .

[0101] In operation S110 , the host device 1100 may provide a first request RQ1 indicating an allocation operation of a unit partition area to the storage device 1200 .

[0102] In some embodiments, the host device 1100 may implement a unit partition size. For example, the first request RQ1 may include a unit partition size. The unit partition size may refer to the number of memory blocks corresponding to one unit partition area.

[0103] In operation S111, the storage device 1200 may perform an allocation operation of a unit partition area based on the first request RQ1. Each unit partition area may have a unit partition size. Each of the plurality of unit partition areas may have a corresponding UZR index value. At least one memory block corresponding to the unit partition area may store data sequentially. After completing the allocation operation, the storage device 1200 may provide a completion response corresponding to the first request RQ1 to the host device 1100.

[0104] In operation S120, the host device 1100 may provide a second request RQ2 indicating an allocation operation of a partition namespace to the storage device 1200. In some embodiments, the second request RQ2 may indicate an allocation operation of a plurality of partition namespaces in a multi-ZNS environment.

[0105] In operation S121, the storage device 1200 may determine first metadata based on the second request RQ2. The first metadata may include a maximum UZR index value for identifying the entire size of the partition namespace. The first metadata may include a mapping ratio obtained by dividing the partition size of the partition namespace by the unit partition size of the unit partition area.

[0106] In operation S122 , the storage device 1200 may store the first metadata in a partition metadata table existing in the storage device 1200 .

[0107] In operation S130, the electronic device 1000 may perform a first user operation. For example, the host device 1100 may provide a request indicating the first user operation to the storage device 1200, the storage device 1200 may perform the first user operation, and the storage device 1200 may provide a completion response of the first user operation to the host device 1100.

[0108] In some embodiments, in operation S120, the first user operation may include at least one of a state conversion operation, a read operation, a write operation, and a metadata read operation of a partition namespace. The state conversion operation may refer to an operation of changing a state according to the ZNS standard (e.g., an operation of converting from an inactive state to an open state). The read operation may refer to an operation of reading data stored in a memory block corresponding to a partition namespace. The write operation may refer to an operation of subsequently performing a sequential write with respect to a memory block corresponding to a partition namespace. The metadata read operation may refer to an operation of reading metadata stored in a partition metadata table.

[0109] In some embodiments, the electronic device 1000 may perform a first user operation and may update the partition metadata table. For example, the first user operation may be a state conversion operation. The electronic device 1000 may store metadata about the changed state in the partition metadata table. As another example, the first user operation may be a write operation. After performing sequential writes, the electronic device 1000 may store metadata about the changed write pointer in the partition metadata table. When a write operation is performed relative to a unit partition area that is not in an open state, the electronic device 1000 may also store metadata indicating the ZSIO state in the partition metadata table.

[0110] In operation S140, the host device 1100 may provide a third request RQ3 indicating a reallocation operation to the storage device 1200. In some embodiments, the reallocation operation may refer to an operation of changing the size of the partition namespace allocated in operation S120, the size of the partition region in the partition namespace, or the number of partition regions in the partition namespace.

[0111] In operation S141, the storage device 1200 may determine second metadata based on the third request RQ3. The second metadata may include a maximum UZR index value for identifying the entire size of the reallocated partition namespace. The second metadata may include a mapping ratio obtained by dividing the partition size of the reallocated partition namespace by the unit partition size of the unit partition area.

[0112] In operation S142, the storage device 1200 may update the partition metadata table so that the first metadata is changed to the second metadata. In detail, the storage device 1200 may search the partition metadata table for the first metadata by using the ZNS index value corresponding to the reallocated partition namespace, and may determine whether the found first metadata matches the second metadata; when it is determined that the found first metadata does not match the second metadata, the storage device 1200 may update the partition metadata table so that the first metadata is changed to the second metadata. When the found first metadata matches the second metadata, the storage device 1200 may maintain the first metadata in the partition metadata table.

[0113] In operation S150, the electronic device 1000 may perform a second user operation. For example, the host device 1100 may provide a request indicating the second user operation to the storage device 1200, the storage device 1200 may perform the second user operation, and the storage device 1200 may provide a completion response of the second user operation to the host device 1100.

[0114] In some implementations, the second user operation may include at least one of a state conversion operation, a read operation, a write operation, and a metadata read operation of the partition namespace reallocated in operation S140.

[0115] In some implementations, the electronic device 1000 may perform a second user operation and may update the partition metadata table. For example, the second user operation may be a state conversion operation or a write operation.

[0116] Fig.10 Describes some embodiments of the present disclosure Fig. 9 Schematic diagram of the reallocation operation. Fig. 9 and Fig.10The partition namespace, the partition area, and the logical area of ​​the unit partition area associated with the reallocation operation performed in operation S140 are described. The host side may correspond to the host device 1100. The storage side may correspond to the storage device 1200.

[0117] Referring to the logical area before performing the reallocation operation, the host side can use the first partition namespace ZNS1, the second partition namespace ZNS2 and the third partition namespace ZNS3. In order to better understand the present disclosure, the logical areas corresponding to the first partition namespace ZNS1, the second partition namespace ZNS2 and the third partition namespace ZNS3 are marked with different shading styles.

[0118] The first partition namespace ZNS1, the second partition namespace ZNS2, and the third partition namespace ZNS3 may be in a logical sequence. For example, the last partition area in the partition areas of the first partition namespace ZNS1 may be in a logical sequence with the first partition area in the partition areas of the second partition namespace ZNS2. The last partition area in the partition areas of the second partition namespace ZNS2 may be in a logical sequence with the first partition area in the partition areas of the third partition namespace ZNS3.

[0119] The first partition namespace ZNS1 may include partition areas ZR11, ZR12, and ZR13. The partition size of the first partition namespace ZNS1 may be twice the unit partition size. The mapping ratio of the first partition namespace ZNS1 may be 2. The maximum UZR index value of the first partition namespace ZNS1 may correspond to UZR6. The entire size of the first partition namespace ZNS1 may be identified by the maximum UZR index value of the first partition namespace ZNS1. Based on the mapping ratio and the maximum UZR index value, the first partition namespace ZNS1 may be mapped to unit partition areas UZR1 to UZR6.

[0120] The second partition namespace ZNS2 may include partition areas ZR21 and ZR22. The partition size of the second partition namespace ZNS2 may be the same as the unit partition size. The mapping ratio of the second partition namespace ZNS2 may be 1. The maximum UZR index value of the second partition namespace ZNS2 may correspond to UZR8. The entire size of the second partition namespace ZNS2 may be identified by the difference between the maximum UZR index value of the second partition namespace ZNS2 and the maximum UZR index value of the first partition namespace ZNS1. Based on the mapping ratio and the maximum UZR index value, the second partition namespace ZNS2 may be mapped to unit partition areas UZR7 and UZR8.

[0121] The third partition namespace ZNS3 may include partition areas ZR31, ZR32, ZR33, ZR34, ZR35, ZR36, ZR37, and ZR38. The partition size of the third partition namespace ZNS3 may be the same as the unit partition size. The mapping ratio of the third partition namespace ZNS3 may be 1. The maximum UZR index value of the third partition namespace ZNS3 may correspond to UZR16. The entire size of the third partition namespace ZNS3 may be identified by the difference between the maximum UZR index value of the third partition namespace ZNS3 and the maximum UZR index value of the second partition namespace ZNS2. Based on the mapping ratio and the maximum UZR index value, the third partition namespace ZNS3 may be mapped to unit partition areas UZR9, UZR10, UZR11, UZR12, UZR13, UZR14, UZR15, and UZR16.

[0122] Next, referring to the logical area after performing the reallocation operation, the reallocation operation may include an operation of reducing the overall size of the first partition namespace ZNS1, an operation of increasing the overall size of the second partition namespace ZNS2, an operation of reducing the overall size of the third partition namespace ZNS3, an operation of changing the mapping ratio of the third partition namespace ZNS3, and an operation of adding (e.g., newly adding)

[0123] Operations in the fourth zone namespace ZNS4.

[0124] After the reallocation operation, the host side may use the first partition namespace ZNS1, the second partition namespace ZNS2, the third partition namespace ZNS3, and the fourth partition namespace ZNS4. For better understanding of the present disclosure, the logical areas corresponding to the reallocated partition namespaces ZNS1, ZNS2, ZNS3, and ZNS4 are marked with different shading styles. The partition namespaces ZNS1, ZNS2, ZNS3, and ZNS4 may be in a logical order.

[0125] The reallocated first partition namespace ZNS1 may include partition areas ZR11 and ZR12. A mapping ratio of the reallocated first partition namespace ZNS1 may be 2. A maximum UZR index value of the reallocated first partition namespace ZNS1 may correspond to UZR4. Based on the mapping ratio and the maximum UZR index value, the reallocated first partition namespace ZNS1 may be mapped to unit partition areas UZR1 to UZR4.

[0126] The reallocated second partition namespace ZNS2 may include partition areas ZR21, ZR22, ZR23, and ZR24. A mapping ratio of the reallocated second namespace ZNS2 may be 1. A maximum UZR index value of the reallocated second partition namespace ZNS2 may correspond to UZR8. Based on the mapping ratio and the maximum UZR index value, the reallocated second partition namespace ZNS2 may be mapped to unit partition areas UZR5, UZR6, UZR7, and UZR8.

[0127] The reallocated third partition namespace ZNS3 may include partition areas ZR31 and ZR32. A mapping ratio of the reallocated partition namespace ZNS3 may be 3. A maximum UZR index value of the reallocated third partition namespace ZNS3 may correspond to UZR14. Based on the mapping ratio and the maximum UZR index value, the reallocated third partition namespace ZNS3 may be mapped to unit partition areas UZR9, UZR10, UZR11, UZR12, UZR13, and UZR14.

[0128] A fourth partition namespace ZNS4 may be newly added through a reallocation operation. The fourth partition namespace ZNS4 may include partition areas ZR41 and ZR42. A mapping ratio of the fourth partition namespace ZNS4 may be 1. A maximum UZR index value of the fourth partition namespace ZNS4 may correspond to UZR16. Based on the mapping ratio and the maximum UZR index value, the fourth partition namespace ZNS4 may be mapped to unit partition areas UZR15 and UZR16.

[0129] In some embodiments, the electronic device may divide the partition namespace by a reallocation operation. For example, the reallocation operation may include an operation of reducing the entire size of the third partition namespace ZNS3 and an operation of adding a fourth partition namespace ZNS4. The entire size of the third partition namespace ZNS3 before the reallocation operation may be equal to the sum of the entire size of the third partition namespace ZNS3 reduced after the reallocation operation and the entire size of the fourth partition namespace ZNS4 added after the reallocation operation. The third partition namespace ZNS3 may be divided by updating the partition metadata table so that the maximum UZR index value of the third partition namespace ZNS3 before the reallocation operation is changed to the maximum UZR index value of the third partition namespace ZNS3 and the maximum UZR index value of the fourth partition namespace ZNS4 after the reallocation operation.

[0130] Fig.11 Describes some embodiments of the present disclosure Fig. 9 Schematic diagram of the reallocation operation. Fig. 9 , Fig.10 and Fig.11 Describes how to update the metadata of the partition metadata table 1213 in the reallocation operation performed in operation S140. The storage device 1200 may include a partition metadata table 1213. The partition metadata table 1213 may include a namespace size table 1213a, a partition size table 1213b, and a partition status table 1213c. The namespace size table 1213a may store a set of ZNS index values ​​and maximum UZR index values. The partition size table 1213b may store a set of ZNS index values ​​and mapping ratios. The partition status table 1213c may store a set of UZR index values, status, and write pointers.

[0131] Referring to the partition metadata table 1213 before performing the reallocation operation, the namespace size table 1213a may store maximum UZR index values ​​UZR6, UZR8, and UZR16 corresponding to the partition namespaces ZNS1, ZNS2, and ZNS3, respectively. The partition size table 1213b may store mapping ratios 2, 1, and 1 corresponding to the partition namespaces ZNS1, ZNS2, and ZNS3, respectively. The partition status table 1213c may store the UZR index value, status, and write pointer of each of the unit partition regions UZR1 to UZR16.

[0132] Referring to the partition metadata table 1213 after performing the reallocation operation, the namespace size table 1213a may store maximum UZR index values ​​UZR4, UZR8, UZR14, and UZR16 corresponding to the partition namespaces ZNS1, ZNS2, ZNS3, and ZNS4, respectively. The partition size table 1213b may store mapping ratios 2, 1, 3, and 1 corresponding to the partition namespaces ZNS1, ZNS2, ZNS3, and ZNS4, respectively. The metadata of the partition state table 1213c may be the same as the metadata before the reallocation operation. The metadata of the partition state table 1213c may be updated by a user operation (e.g., a state conversion operation or a write operation).

[0133] As described above, according to an embodiment of the present disclosure, the electronic device 1000 can dynamically change the size of a partition namespace or a partition area or the number of partition namespaces or partition areas by only updating the metadata of the partition namespace in the partition metadata table 1213, while maintaining the mapping relationship between the memory block and the unit partition area without modification. Figure 3 Compared with a conventional electronic device ED, resources, power, and processing time of the electronic device 1000 may be reduced, which is desirable for changing the size of a partition namespace or a partition area or the number of partition namespaces or partition areas.

[0134] Fig.12 is a flowchart describing a method of operating a storage device according to some embodiments of the present disclosure. Fig.12 The storage device communicates with the host device. The storage device may include a partition metadata table.

[0135] In operation S210, the storage device allocates a plurality of unit partition areas, each unit partition area having a unit partition size. The plurality of unit partition areas may be mapped to at least one memory block in the nonvolatile memory device. The unit partition size may refer to the number of memory blocks mapped to one unit partition area. The storage device may internally manage the unit partition area, and the host device may have difficulty in directly identifying the unit partition area.

[0136] In operation S220, the storage device receives a request indicating an allocation operation of a target partition namespace from a host device. The target partition namespace may include a plurality of partition areas. The partition areas may also be referred to as "host partition areas". Each partition area may have a target partition size.

[0137] In operation S230, the storage device determines a target mapping ratio by dividing a target partition size of a target partition namespace by a unit partition size based on the request in operation S220. The target mapping ratio may indicate the number of unit partition regions mapped to one host partition unit (eg, partition region).

[0138] In operation S240, the storage device stores the target mapping ratio in the partition metadata table. Thereafter, the storage device may provide a completion response to the host device.

[0139] Fig.13 is a flowchart describing a method of operating a storage device according to some embodiments of the present disclosure. Fig.13 The storage device communicates with the host device. The storage device may include a partition metadata table.

[0140] In operation S310, the storage device allocates a plurality of unit partition areas, each unit partition area having a corresponding UZR index value. The plurality of unit partition areas may have the same size. The UZR index value may be a unique value for identifying a corresponding unit partition area among the plurality of unit partition areas.

[0141] In operation S320, the storage device receives a request indicating an allocation operation of a first partition namespace and a second partition namespace. The first partition namespace may include a plurality of first partition areas. The first partition area of ​​the first partition namespace may be mapped to some of the plurality of unit partition areas allocated in operation S310. The second partition namespace may include a plurality of second partition areas. The second partition area of ​​the second partition namespace may be mapped to other unit partition areas of the plurality of unit partition areas allocated in operation S310.

[0142] In operation S331, the storage device determines a first maximum UZR index value of the first partition namespace based on the request received in operation S320. The first maximum UZR index value may correspond to a last unit partition area among unit partition areas corresponding to the first partition namespace.

[0143] In operation S332, the storage device determines a second maximum UZR index value of the second partition namespace based on the request received in operation S320. The second maximum UZR index value may correspond to a last unit partition area among unit partition areas corresponding to the second partition namespace.

[0144] In operation S340, the storage device stores the first maximum UZR index value and the second maximum UZR index value in the partition metadata table. Thereafter, the storage device may provide a completion response to the host device.

[0145] Fig.14 is a flowchart describing a method of operating a storage device according to some embodiments of the present disclosure. Fig.14 The storage device communicates with the host device. The storage device may include a partition metadata table.

[0146] In operation S410, the storage device allocates a plurality of unit partition areas. Each unit partition area may have a unit partition size. Each of the plurality of unit partition areas may have a corresponding UZR index value. Operation S410 may correspond to Fig.12 Operation S210 and Fig.13 Operation S310.

[0147] In operation S420, the storage device performs an allocation operation of the target partition namespace. The target partition namespace may include a plurality of partition areas. The plurality of partition areas of the target partition namespace may be mapped to the plurality of unit partition areas allocated in operation S410. The storage device may store a mapping ratio of the target partition namespace and a maximum UZR index value of the target partition namespace in a partition metadata table. The storage device may provide a completion response to the host device. Operation S420 may correspond to Fig.12 Operation S220, operation S230 and operation S240 and Fig.13 Operation S320, operation S331, operation S332 and operation S340.

[0148] In operation S430, the storage device receives a request for a metadata read operation of a target partition area of ​​a target partition namespace from a host device. The target partition area may be one of a plurality of partition areas included in the target partition namespace. A unit partition area corresponding to the target partition area may have a status and a write pointer, and the status and the write pointer may be managed by using a partition metadata table.

[0149] In operation S440, the storage device obtains first metadata of a unit partition area corresponding to a target partition area from a partition metadata table. The first metadata may include a state or a write pointer of the corresponding unit partition area.

[0150] In operation S450, the storage device determines second metadata of the target partition area based on the first metadata obtained in operation S440. The second metadata may include a state or a write pointer of the target partition area. That is, the storage device may collect metadata of the unit partition area corresponding to the target partition area to determine the metadata of the target partition area.

[0151] In some embodiments, the storage device determines second metadata about the write pointer of the target partition area based on the first metadata about the write pointer of the unit partition area. For example, the storage device may determine that the write pointer of the target partition area corresponds to the sum of the write pointers of the unit partition areas corresponding to the target partition area.

[0152] In some embodiments, the storage device determines second metadata about the state of the target partition area based on the first metadata about the state of the unit partition area. For example, the storage device may determine whether there is an open state in the state of the unit partition area corresponding to the target partition area. In response to determining that there is an open state in the state of the unit partition area corresponding to the target partition area, the storage device may determine that the state of the target partition area is an open state.

[0153] In operation S460, the storage device provides the host device with second metadata. The second metadata may be a response to the metadata read operation of the target partition area performed in operation S430.

[0154] According to an embodiment of the present disclosure, a storage device for allocating a unit partition area and an operating method thereof are provided.

[0155] In addition, a storage device and an operating method thereof are provided, which, when a reallocation request for a host partition area is received from a host side, reduces the resources, power, and processing time required for changing the size of a namespace or a host partition area by only updating metadata between a unit partition area and a host partition area without deallocation or reallocation of physical memory blocks.

[0156] Although the present disclosure includes many specific implementation details, these should not be interpreted as limiting the scope of the claimed protection. Certain features described in the context of separate implementations in the present disclosure may also be implemented in combination in a single implementation. On the contrary, the various features described in the context of a single implementation may also be implemented in multiple implementations individually or in any suitable sub-combination. In addition, although the features may be described above as working in certain combinations and even initially claimed as such, one or more features from the combination may be deleted from the combination in some cases, and the combination may be directed to a sub-combination or a variation of the sub-combination.

[0157] While the present disclosure has been described with reference to the embodiments thereof, it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the present disclosure as set forth in the appended claims.

Claims

1. A method for operating a storage device, the method include: allocating a plurality of unit partition areas, each unit partition area having a unit partition size, wherein each unit partition area in the plurality of unit partition areas is mapped to at least one corresponding memory block in a plurality of memory blocks in a storage device, and wherein each unit partition area in the plurality of unit partition areas supports sequential writes in accordance with a zoned name space (ZNS) standard; receiving, from a host device, a first request indicating an allocation operation of a first partitioned namespace conforming to a ZNS standard; Based on the first request, determining a first mapping ratio by dividing a first partition size of the first partition namespace by a unit partition size; as well as The first mapping ratio is stored in a partition metadata table of the storage device.

2. The method according to claim 1, in, The partition metadata table includes: a namespace size table configured to store a first maximum unit partition region (UZR) index value corresponding to a last unit partition region among the unit partition regions corresponding to the first partition namespace among the plurality of unit partition regions; a partition size table configured to store a first mapping ratio; and The partition status table is configured to store a status and a write pointer of each unit partition area among the plurality of unit partition areas.

3. The method according to claim 1, in, The unit partition size corresponds to one of the following: Memory blocks; a first number of memory blocks that conform to an operating policy of the memory device; or A second number of memory blocks is determined by the host device.

4. The method according to claim 1, in, Each of the plurality of unit partition areas has a corresponding UZR index value, and Wherein, the method further comprises: determining a first maximum UZR index value for the first partition namespace based on the first request, wherein the first maximum UZR index value corresponds to a last unit partition area among the unit partition areas corresponding to the first partition namespace among the plurality of unit partition areas; and The first maximum UZR index value is stored in a partition metadata table of the storage device.

5. The method of claim 1 , wherein the allocation operation comprises an allocation operation of a second partitioned namespace that complies with a ZNS standard, and Wherein a last partition region among the partition regions of the first partition namespace is logically sequential with a first partition region among the partition regions of the second partition namespace.

6. The method according to claim 5, in, Each of the plurality of unit partition areas has a corresponding UZR index value, and Wherein, the method further comprises: determining, based on the first request, a second mapping ratio by dividing a second partition size of the second partition namespace by the unit partition size; storing the second mapping ratio in a partition metadata table of the storage device; determining a second maximum UZR index value for the second partition namespace based on the first request, wherein the second maximum UZR index value corresponds to a last unit partition area among the plurality of unit partition areas corresponding to the second partition namespace; and The second largest UZR index value is stored in a partition metadata table of the storage device.

7. The method according to claim 1, further comprising: include: After storing the first mapping ratio, receiving a second request indicating a first user operation of the first partition namespace from the host device, wherein the first user operation indicates a state conversion operation, a read operation, a write operation, or a metadata read operation of the first partition namespace; as well as The first user operation is performed based on the second request.

8. The method according to claim 7, in, The first user operation indicates a state conversion operation or a write operation of a target partition area in the partition area of ​​the first partition namespace, The performing of the first user operation based on the second request includes: updating the partition metadata table so that the write pointer or status of each unit partition area corresponding to the target partition area is changed, and The status complies with the ZNS standard.

9. The method according to claim 7, in, The first user operation indicates a metadata read operation of a target partition region in the partition regions of the first partition namespace, and The performing of the first user operation based on the second request includes: obtaining, from the partition metadata table, first metadata of a unit partition area corresponding to a target partition area among the plurality of unit partition areas; determining second metadata of the target partition area based on the first metadata; and The second metadata is provided to the host device.

10. The method according to claim 9, in, The first metadata includes first write pointers respectively corresponding to unit partition areas corresponding to the target partition areas, wherein the second metadata includes a second write pointer to the target partition area, and The step of determining the second metadata of the target partition area based on the first metadata includes: It is determined that the second write pointer of the target partition area is the sum of the first write pointers of the unit partition areas corresponding to the target partition area.

11. The method according to claim 9, in, The first metadata includes first states respectively corresponding to unit partition areas corresponding to the target partition areas, the first states complying with the ZNS standard, wherein the second metadata includes a second state of the target partition area, the second state conforming to the ZNS standard, and The step of determining the second metadata of the target partition area based on the first metadata includes: determining whether an open state exists in the first state; and In response to determining that the open state exists in the first state, the second state is determined to be the open state.

12. The method according to claim 1, further comprising: include: After storing the first mapping ratio, receiving a third request from the host device indicating a reallocation operation of the first partitioned namespace that complies with the ZNS standard; Based on the third request, determining a third mapping ratio by dividing a third partition size of the first partition namespace by the unit partition size; and A third mapping ratio in a partition metadata table of the storage device is updated.

13. The method according to claim 12, in, Updating the third mapping ratio in the partition metadata table of the storage device includes: searching a partition metadata table for a first mapping ratio by using a ZNS index value corresponding to the first partition namespace; determining whether the third mapping ratio is consistent with the first mapping ratio; and In response to determining that the third mapping ratio is inconsistent with the first mapping ratio, the partition metadata table is updated so that the first mapping ratio is changed to the third mapping ratio.

14. The method according to claim 12, wherein the reallocation operation of the first partition namespace comprises an operation of reducing the first partition namespace and an operation of adding a third partition namespace, in, Before the reallocation operation, the first logical size of the first partition namespace is equal to the sum of the second logical size of the first partition namespace reduced thereby and the third logical size of the third partition namespace, and Wherein, the method further comprises: determining, based on the third request, a third maximum UZR index value corresponding to the first partition namespace thereby reduced; Based on the third request, determining a fourth maximum UZR index value corresponding to a third partition namespace; and A partition metadata table of the storage device is updated so that a previous UZR index value of the first partition namespace before the reallocation operation is changed to a third maximum UZR index value and a fourth maximum UZR index value.

15. The method according to claim 12, further comprising: include: After updating the third mapping ratio, receiving a fourth request indicating a second user operation of the first partition namespace from the host device, wherein the second user operation indicates a state conversion operation, a read operation, a write operation, or a metadata read operation of the first partition namespace; as well as A second user operation is performed based on the fourth request.

16. A method of operating a storage device in communication with a host device, the method include: Allocating a plurality of unit zoned regions, each unit zoned region having a corresponding unit zoned region (UZR) index value; receiving a request indicating an allocation operation of a first partitioned namespace and a second partitioned namespace that conforms to a zoned namespace (ZNS) standard; determining a first maximum UZR index value for the first partition namespace based on the request, wherein the first maximum UZR index value corresponds to a last unit partition area among the unit partition areas corresponding to the first partition namespace among the plurality of unit partition areas; determining a second maximum UZR index value for the second partition namespace based on the request, wherein the second maximum UZR index value corresponds to a last unit partition area among the unit partition areas corresponding to the second partition namespace among the plurality of unit partition areas; as well as The first maximum UZR index value and the second maximum UZR index value are stored in a partition metadata table of the storage device.

17. The method according to claim 16, in, Each of the plurality of unit partition areas has a unit partition size, wherein each partition region of the first partition namespace has a first partition size, wherein each partition region of the second partition namespace has a second partition size, and Wherein, the method further comprises: Based on the request, determining a first mapping ratio by dividing a first partition size of the first partition namespace by the unit partition size; storing the first mapping ratio in a partition metadata table of the storage device; Based on the request, determining a second mapping ratio by dividing a second partition size of the second partition namespace by the unit partition size; and The second mapping ratio is stored in a partition metadata table of the storage device.

18. A storage device, include: a command manager configured to receive, from a host device, a first request indicating an allocation operation of a first zoned namespace conforming to a zoned namespace (ZNS) standard, and configured to generate a second request indicating the allocation operation; a partition manager configured to allocate a plurality of unit partition regions, each unit partition region having a unit partition size and a corresponding unit partition region (UZR) index value; Partition metadata table; as well as a nonvolatile memory device including a plurality of memory blocks corresponding to the plurality of unit partition areas, Wherein, the partition manager is further configured to: Based on the second request, determining a first mapping ratio by dividing a first partition size of the first partition namespace by a unit partition size; storing the first mapping ratio in a partition metadata table; Based on the second request, determining a first maximum UZR index value corresponding to a last unit partition area among the unit partition areas corresponding to the first partition namespace among the plurality of unit partition areas; and The first largest UZR index value is stored in the partition metadata table.

19. The storage device of claim 18, wherein the allocation operation comprises an allocation operation of a second partition namespace that complies with a ZNS standard, and in, The partition manager is also configured to: determining, based on the second request, a second mapping ratio by dividing a second partition size of the second partition namespace by the unit partition size; storing the second mapping ratio in the partition metadata table; Based on the second request, determining a second maximum UZR index value, the second maximum UZR index value corresponding to a last unit partition area among the unit partition areas corresponding to the second partition namespace among the plurality of unit partition areas; as well as The second largest UZR index value is stored in the partition metadata table.

20. The storage device according to claim 18, in, The partition metadata table includes: a namespace size table configured to store a first maximum UZR index value; a partition size table configured to store a first mapping ratio; and The partition status table is configured to store a status and a write pointer of each unit partition area among the plurality of unit partition areas.